Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Investigation of the potential application of rhenium in medical imaging De La Vega, José Carlos 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2017_november_delavega_josé.pdf [ 3.51MB ]
Metadata
JSON: 24-1.0354674.json
JSON-LD: 24-1.0354674-ld.json
RDF/XML (Pretty): 24-1.0354674-rdf.xml
RDF/JSON: 24-1.0354674-rdf.json
Turtle: 24-1.0354674-turtle.txt
N-Triples: 24-1.0354674-rdf-ntriples.txt
Original Record: 24-1.0354674-source.json
Full Text
24-1.0354674-fulltext.txt
Citation
24-1.0354674.ris

Full Text

  INVESTIGATION OF THE POTENTIAL APPLICATION OF RHENIUM IN MEDICAL IMAGING  by  José Carlos De La Vega B.Sc., Instituto Tecnológico y de Estudios Superiores de Monterrey, 2012  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF  THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES   (Pharmaceutical Sciences)  THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)  August 2017  © José Carlos De La Vega, 2017 ii  Abstract  Many thermodynamically stable coordination compounds have been synthesized using rhenium, a chemically versatile transition metal. Some of these rhenium complexes have been studied for utilization in nuclear medicine. However, many of their applications in medical imaging remain relatively unexplored and require further investigation.  A comprehensive study was conducted in preclinical and clinical X-ray equipment to examine the use of rhenium as a contrast agent for X-ray imaging. Usually, there is a trade-off between image quality and radiation dose. This experimental work, along with theoretical Monte Carlo calculations, showed that it is feasible to preserve image quality and minimize radiation dose simultaneously when a rhenium-based formulation is utilized. This research provided thorough evidence of rhenium’s usefulness in X-ray imaging. Another application of rhenium was evaluated by producing a radiopaque, biodegradable electrospun scaffold containing a rhenium complex. Typically, catheterizations are performed under X-ray imaging guidance, but most catheters are radiolucent. After coating catheters with this scaffold, they became strongly radiopaque. Even a thin rhenium-doped coating has the potential of enhancing the contrast during catheterizations, which might be helpful in placing catheters more rapidly and precisely.  Not only large medical devices, but also microsized carriers can be made radiopaque. An issue with embolic microspheres is their lack of contrast. To improve their visibility, potentially toxic contrast agents are co-administered in X-ray imaging-guided embolotherapy. Using a microfluidic technology, radiopaque, biodegradable microspheres made of a custom-synthesized polymer containing a rhenium complex were produced. Upon increasing the polymer’s rhenium concentration, these microspheres could be utilized in embolotherapy.  Rhenium’s radioisotope 188Re is a mixed 𝛽− and 𝛾 emitter and can thus be exploited in another imaging modality: single photon emission computed tomography (SPECT). The biodistribution of microspheres labeled with 188Re was evaluated in a hepatocellular carcinoma-bearing rat model. Although challenging in clinical practice, the radiation doses to the tumor and the healthy liver tissue were calculated. The radiation dose from the 𝛽− emissions yields these “imageable” microspheres theranostic, with quantifiable cancer radiotherapeutic potential.  This work established the foundations to guide further research on the development of biodegradable devices doped with rhenium for medical imaging.  iii  Lay Summary  This study aimed to provide evidence of the usefulness of rhenium, a transition metal, in medical imaging.  We showed that rhenium is fully visible in X-ray images. This property was exploited by making a rhenium-doped coating for catheters. A thin layer of this material made the catheters visible under X-rays, which has the potential of helping physicians in moving them inside the body.  Also visible under X-rays, microspheres, which are tiny sub-millimeter beads, were produced using a rhenium-doped polymer. These “imageable” microspheres could be used in embolotherapy, a medical procedure where microspheres are administered into patients to block a tumor’s blood vessels.  Similar microspheres were then bound to radioactive rhenium. The radiation emitted by these microspheres can kill cancer cells, but it can also be used to track the location of the microspheres in the body.  Through this work, new applications of rhenium were successfully identified.      iv  Preface  I designed, conducted, and interpreted the experiments included in this dissertation. Dr. Urs Hafeli, my supervisor, provided advice in the execution of the experiments as well as in the preparation of this dissertation. The contributions of our collaborators are outlined below.    Chapter 2  Dr. Pedro Esquinas performed the Monte Carlo calculations. I executed all other experiments. Dr. Bradford Gill assisted with the Small Animal Radiation Research Platform (SARRP). Dr. Yogesh Thakur provided help with the X-ray camera for digital radiography.  The micro-computed tomography (𝜇CT) study was conducted at the Centre for High-Throughput Phenogenomics (CHTP) at the University of British Columbia (UBC), a facility supported by the Faculty of Dentistry. Imaging with the SARRP was carried out in the Department of Medical Physics at the BC Cancer Agency. The digital radiography study was performed in the Department of X-Ray/Radiology at the Leslie Diamond Health Care Centre, an institute within the Vancouver General Hospital (VGH).  Chapter 3  Dr. Katayoun Saatchi synthesized and characterized the rhenium phosphinophenolate complex. Samples of the scaffolds were sent to Exova in Surrey, Canada for analysis of the content of rhenium via inductively coupled plasma mass spectroscopy (ICP-MS). I performed the rest of the experiments with assistance of my undergraduate student, Jovan Gill. The majority of the experiments were executed at the UBC’s Faculty of Pharmaceutical Sciences. The only exceptions were microscopy of the scaffolds and imaging of the catheters by 𝜇CT, which were both conducted at CHTP.   Chapter 4  Dr. Katayoun Saatchi synthesized the polymer functionalized with rhenium and labeled the microspheres with 188Re. Rhonda Hildebrandt performed the hepatic intra-arterial catheterizations under the clinical supervision of Dr. Laura Mowbray. Dr. Pedro Esquinas carried out the Monte Carlo calculations. Dr. Igor Moskalev assisted in some of the ultrasound-guided injections, but I completed most of the procedures alone. I executed all the other experiments. v  Imaging by 𝜇CT was performed at CHTP. The viscosity measurements were conducted at the UBC’s Department of Materials Engineering. The osmolality measurements were carried out at the Centre for Drug Research and Development (CDRD). The radioembolization study was performed at the Centre for Comparative Medicine (CCM). All other experiments were conducted at the UBC’s Faculty of Pharmaceutical Sciences.  At the time of writing, extended versions of the following sections have been published as review articles:  Section 1.3: De La Vega JC and Häfeli UO (2015). Utilization of Nanoparticles as X-Ray Contrast Agents for Diagnostic Imaging Applications. Contrast Media and Molecular Imaging. 10: 81-95.   Section 4.1.2: De La Vega JC, Elischer P, Schneider T, and Häfeli UO (2013). Uniform Polymer Microspheres: Monodispersity Criteria, Methods of Formation and Applications. Nanomedicine. 8: 265-285.  Before publication of this dissertation, modified versions of some sections will be submitted as manuscripts as follows:  Chapter 2: De La Vega JC, Gill JK, Esquinas PL, Jessa S, Gill B, Thakur Y, Saatchi K, and Häfeli UO. Comparison of the Contrast-to-Noise Ratio and the Relative Absorbed Dose of Rhenium and Iodine in Preclinical and Clinical X-Ray Imaging.   Chapter 3: De La Vega JC, Gill JK, Peters JM, Jessa S, Saatchi K, and Häfeli UO. Electrospun Biodegradable Rhenium-Doped Scaffolds as a Radiopaque Coating for Catheters.   Furthermore, part of Chapter 4 will be submitted as a short communication:  Section 4.2.2/4.3.2: De La Vega JC, Esquinas PL, Rodríguez-Rodríguez C, Nosrati Z, Bokharaei M, Moskalev I, Thakor AS, Liu D, Saatchi K, Celler A, and Häfeli UO. Radioembolization of Biodegradable, Uniformly-Sized 188Re-Labeled Microspheres in an HCC-Bearing Rat Model.   All studies requiring the utilization of animals and biohazards were reviewed and approved by the Animal Care Committee and the Biosafety Committee at UBC, respectively. The following two certificates were granted: A15-0244 and B14-0187.  vi  Table of Contents  Abstract .......................................................................................................................................... ii Lay Summary ............................................................................................................................... iii Preface ........................................................................................................................................... iv Table of Contents ......................................................................................................................... vi List of Tables ................................................................................................................................ ix List of Figures ................................................................................................................................ x List of Abbreviations ................................................................................................................. xiii Acknowledgements .................................................................................................................... xvi Dedication ................................................................................................................................. xviii Chapter 1: Introduction ............................................................................................................... 1 1.1 Overview of the Dissertation ........................................................................................... 1 1.2 Chemical and Radioactive Properties of Rhenium .......................................................... 3 1.3 Medical Imaging and X-Ray Contrast Agents ................................................................. 4 1.3.1 Development of X-Ray Contrast Agents with High-𝑍 Elements ............................. 7 Chapter 2: Investigation of the X-Ray Attenuation of Rhenium ............................................. 9 2.1 Background ...................................................................................................................... 9 2.1.1 X-Ray Production and Interaction with Matter ........................................................ 9 2.1.2 Assessment of Image Quality and Radiation Dose ................................................. 15 2.2 Materials and Methods ................................................................................................... 17 2.2.1 Preparation of Rhenium- and Iodine-Based Solutions............................................ 18 2.2.2 Imaging of Rhenium- and Iodine-Based Solutions ................................................. 18 2.2.2.1 Micro-Computed Tomography ........................................................................... 18 2.2.2.2 Cone-Beam Computed Tomography .................................................................. 21 2.2.2.3 Planar X-Ray Imaging ......................................................................................... 22 2.2.2.4 Digital Radiography ............................................................................................ 25 2.3 Results and Discussion ................................................................................................... 27 2.3.1 Imaging of Rhenium- and Iodine-Based Solutions ................................................. 28 2.3.1.1 Micro-Computed Tomography ........................................................................... 28 2.3.1.2 Cone Beam Computed Tomography ................................................................... 35 2.3.1.3 Planar X-Ray Imaging ......................................................................................... 37 vii  2.3.1.4 Digital Radiography ............................................................................................ 38 2.4 Conclusions .................................................................................................................... 44 Chapter 3: Development and Imaging of Radiopaque Electrospun Scaffolds ...................... 46 3.1 Background .................................................................................................................... 46 3.1.1 Principles and Applications of the Electrospinning Technique .............................. 46 3.1.2 X-Ray Imaging Guidance in Catheterizations ........................................................ 49 3.2 Materials and Methods ................................................................................................... 51 3.2.1 Optimization of the Electrospinning Technique ..................................................... 51 3.2.2 Development of a Rhenium-Doped Scaffold by Electrospinning .......................... 53 3.2.2.1 Production and Characterization of the Scaffold ................................................ 54 3.2.2.2 Appraisal of the Cytotoxicity .............................................................................. 54 3.2.3 Coating and X-Ray Imaging of Catheters ............................................................... 56 3.3 Results and Discussion ................................................................................................... 58 3.3.1 Optimization of the Electrospinning Technique ..................................................... 58 3.3.2 Development of a Rhenium-Doped Scaffold by Electrospinning .......................... 64 3.3.2.1 Production and Characterization of the Scaffold ................................................ 64 3.3.2.2 Appraisal of the Cytotoxicity .............................................................................. 66 3.3.3 Coating and X-Ray Imaging of Catheters ............................................................... 68 3.4 Conclusions .................................................................................................................... 71 Chapter 4: Development and Imaging of Radiopaque and Radioactive Microspheres ....... 72 4.1 Background .................................................................................................................... 72 4.1.1 Embolization Therapy ............................................................................................. 73 4.1.1.1 Embolization Therapy for Hepatocellular Carcinoma ........................................ 73 4.1.2 Production of Microspheres by Flow Focusing ...................................................... 78 4.2 Materials and Methods ................................................................................................... 79 4.2.1 Development and Imaging of Radiopaque Microspheres with Rhenium ............... 80 4.2.1.1 Production and Characterization of the Microspheres ........................................ 80    4.2.1.1.1    Production ..................................................................................................... 80    4.2.1.1.2    Characterization ............................................................................................ 82 4.2.1.2 X-Ray Imaging .................................................................................................... 82 4.2.2 Development and Imaging of Radioactive Microspheres with Rhenium-188 ........ 83 viii  4.2.2.1 Animals ............................................................................................................... 84 4.2.2.2 Production, Radiolabeling, and Characterization of the Microspheres ............... 84    4.2.2.2.1    Production ..................................................................................................... 84    4.2.2.2.2    Radiolabeling ................................................................................................ 85    4.2.2.2.3    Characterization ............................................................................................ 85    4.2.2.2.4    Development of a Parenteral Formulation .................................................... 85 4.2.2.3 Development of a Hepatocellular Carcinoma-Bearing Animal Model ............... 86    4.2.2.3.1    Culture and Preparation of Doses of N1-S1 Cells ........................................ 86    4.2.2.3.2    Ultrasound-Guided Inoculation of N1-S1 Cells into the Liver .................... 87 4.2.2.4 Radioembolization and Quantitative SPECT Imaging........................................ 88    4.2.2.4.1    Administration Device .................................................................................. 89    4.2.2.4.2    Hepatic Intra-Arterial Catheterization .......................................................... 91    4.2.2.4.3    Quantitative SPECT Imaging and Dosimetry .............................................. 93 4.3 Results and Discussion ................................................................................................... 94 4.3.1 Development and Imaging of Radiopaque Microspheres with Rhenium ............... 95 4.3.1.1 Production and Characterization of the Microspheres ........................................ 95 4.3.1.2 X-Ray Imaging .................................................................................................... 96 4.3.2 Development and Imaging of Radioactive Microspheres with Rhenium-188 ........ 98 4.3.2.1 Production, Radiolabeling, and Characterization of the Microspheres ............... 98 4.3.2.2 Development of a Hepatocellular Carcinoma-Bearing Animal Model ............. 102 4.3.2.3 Radioembolization and Quantitative SPECT Imaging...................................... 105 4.4 Conclusions .................................................................................................................. 108 Chapter 5: Conclusions ............................................................................................................ 110 5.1 Significance and Contribution of the Research ............................................................ 110 5.2 Strengths and Limitations of the Research ................................................................... 111 5.3 Future Directions .......................................................................................................... 112 References .................................................................................................................................. 115   ix  List of Tables  Chapter 1 Introduction  Table 1.1       Energy of Most Abundant 188Re Emissions .............................................................. 3  Chapter 2 Investigation of the X-Ray Attenuation of Rhenium  Table 2.1       Settings for Micro-Computed Tomography ........................................................... 20 Table 2.2       Settings for Cone Beam Computed Tomography ................................................... 22 Table 2.3       Settings for Planar X-Ray Imaging ......................................................................... 23 Table 2.4       Settings for Digital Radiography ............................................................................ 26 Table 2.5       Linear Regression Analysis .................................................................................... 29 Table 2.6       Measurement of the Air-Kerma and the Mean Energy of the X-Ray Beam .......... 40 Table 2.7       Relative Average Absorbed Dose ........................................................................... 44  Chapter 3 Development and Imaging of Radiopaque Electrospun Scaffolds  Table 3.1       Settings for Electrospinning of PCL-Based Scaffolds ............................................ 53 Table 3.2       Incubation of Samples of Scaffolds for Cytotoxicity Assay................................... 55 Table 3.3       Thickness Appraisal of 45 kDa Polycaprolactone-Based Scaffolds ....................... 59 Table 3.4       Mean Pixel Intensity of the Coatings ...................................................................... 71  Chapter 4 Development and Imaging of Radiopaque and Radioactive Microspheres  Table 4.1       Dose of 188Re-Labeled Microspheres per Rat ......................................................... 89 Table 4.2       Radiation Doses to the Tumor and the Healthy Liver Tissue ............................... 107   x  List of Figures  Chapter 1 Introduction  Figure 1.1      Iodine-Based X-Ray Contrast Agents ...................................................................... 6  Chapter 2 Investigation of the X-Ray Attenuation of Rhenium  Figure 2.1      X-Ray Tube ............................................................................................................ 10 Figure 2.2      Photoelectric Effect and Compton Effect ............................................................... 12 Figure 2.3      X-Ray Beam Attenuation. ...................................................................................... 13 Figure 2.4      Effect of the X-Ray Tube Potential on the X-Ray Spectra .................................... 14 Figure 2.5      Contrast and Noise ................................................................................................. 16 Figure 2.6      Enhanced Resolution Phantom ............................................................................... 19 Figure 2.7      Preclinical Micro-Computed Tomography Scanner ............................................... 20 Figure 2.8      Small Animal Radiation Research Platform........................................................... 22 Figure 2.9      Heel Effect .............................................................................................................. 24 Figure 2.10    Setup for Planar X-Ray Imaging ............................................................................ 24 Figure 2.11    Digital Radiography Camera .................................................................................. 25 Figure 2.12    Sample Holder ........................................................................................................ 25 Figure 2.13    Geometry of Monte Carlo Simulation .................................................................... 27 Figure 2.14    Evaluation of the Contrast-to-Noise Ratio in Micro-Computed Tomography ....... 29 Figure 2.15    Effect of the X-Ray Tube Potential on the Contrast-to-Noise Ratio ...................... 30 Figure 2.16    Effect of the Amount of Additional Filtration on the Contrast-to-Noise Ratio ..... 31 Figure 2.17    Micro-Computed Tomography Imaging of Rhenium and Iodine .......................... 32 Figure 2.18    Effect of the Photon Energy on the Mass X-Ray Attenuation Coefficient ............ 33 Figure 2.19    Effect of Copper on the 120 kVp X-Ray Spectra ................................................... 34 Figure 2.20    Effect of the X-Ray Tube Potential on the Contrast-to-Noise Ratio ...................... 35 Figure 2.21    Effect of the Amount of Additional Filtration on the Contrast-to-Noise Ratio ..... 36 Figure 2.22    Effect of the X-Ray Tube Potential on the Contrast-to-Noise Ratio ...................... 38 Figure 2.23    Evaluation of the Contrast-to-Noise Ratio in Digital Radiography ....................... 39 xi  Figure 2.24    Rhenium-to-Iodine X-Ray Attenuation Ratio ........................................................ 42 Figure 2.25    Depth Relative Absorbed Dose Profiles ................................................................. 44  Chapter 3 Development and Imaging of Radiopaque Electrospun Scaffolds  Figure 3.1      Schematic of the Electrospinning Technique ......................................................... 48 Figure 3.2      Electrospinning Setup ............................................................................................. 52 Figure 3.3      Configurations for the Collection of Scaffolds ...................................................... 52 Figure 3.4      Specimen Micro-Computed Tomography Scanner ................................................ 57 Figure 3.5      Examples of 45 kDa Polycaprolactone-Based Scaffolds ....................................... 59 Figure 3.6      Viscosity of Polycaprolactone-Based Solutions ..................................................... 62 Figure 3.7      Morphology Inspection of Polycaprolactone-Based Scaffolds .............................. 63 Figure 3.8      Rhenium Phosphinophenolate Complex ................................................................ 65 Figure 3.9      Morphology Inspection of Rhenium-Doped and Rhenium-Free Scaffolds ........... 66 Figure 3.10    HEK-293 Cell Confluency Variation Post-Scaffold Treatment ............................. 67 Figure 3.11    HEK-293 Cell Confluency Variation Post-Ammonium Perrhenate Treatment ..... 68 Figure 3.12    X-Ray Imaging of Coated Catheters ...................................................................... 69 Figure 3.13    Pixel Intensity Analysis of the Coatings ................................................................ 70  Chapter 4 Development and Imaging of Radiopaque and Radioactive Microspheres  Figure 4.1      Flow Focusing Microfluidic Chip .......................................................................... 78 Figure 4.2      Ultrasound Machine ............................................................................................... 87 Figure 4.3      Setup for Ultrasound-Guided Inoculation of N1-S1 Cells into the Liver .............. 88 Figure 4.4      Device for Administration of 188Re-Labeled Microspheres ................................... 90 Figure 4.5      Set of Catheters for Administration of 188Re-Labeled Microspheres ..................... 90 Figure 4.6      Schematic of the Hepatic Intra-Arterial Catheterization ........................................ 92 Figure 4.7      VECTor Imaging Platform ..................................................................................... 93 Figure 4.8      Size Distribution and Morphology of the Rhenium-Doped Microspheres ............ 95 Figure 4.9      Structure of Rhenium-Functionalized Polymer ...................................................... 96 Figure 4.10    Micro-Computed Tomography Imaging ................................................................ 97 xii  Figure 4.11    Pixel Intensity Analysis .......................................................................................... 98 Figure 4.12    Size Distribution of the Microspheres .................................................................... 99 Figure 4.13    Morphology Inspection of Microspheres Before and After 188Re Labeling......... 100 Figure 4.14    Schematic of the Surface of the 188Re-Labeled Microspheres ............................. 100 Figure 4.15    Turbidimetric Analysis of ~40 𝜇m Microspheres  .............................................. 102 Figure 4.16    Population Doubling Level of Subcultures of N1-S1 Cells ................................. 104 Figure 4.17    Stages of Ultrasound-Guided Inoculation of N1-S1 Cells into the Liver ............ 105 Figure 4.18    Biodistribution of 188Re-Labeled Microspheres ................................................... 106 Figure 4.19    Effect of Time on the Biodistribution of 188Re-Labeled Microspheres ................ 108    xiii  List of Abbreviations  188Re-MS Microspheres Labeled with 188Re 2D Two-Dimensional 3D Three-Dimensional 90Y-MS Microspheres Labeled with 90Y 99mTc-MAA Macroaggregated Albumin Labeled with 99mTc 99mTc-MS Microspheres Labeled with 99mTc AuNPs Gold Nanoparticles BiNPs Bismuth Nanoparticles C-MS Rhenium-Free Microspheres (i.e., Polymer-Based Microspheres) CBCT Cone Beam Computed Tomography CCAC Canadian Council on Animal Care CCM Centre for Comparative Medicine CHA Common Hepatic Artery CHTP Centre for High-Throughput Phenogenomics 𝐶𝑁𝑅 Contrast-to-Noise Ratio CP Continuous Phase CT Computed Tomography cTACE Conventional Transarterial Chemoembolization 𝐶𝑉 Coefficient of Variation DEB-TACE Drug-Eluting Bead Transarterial Chemoembolization DMEM Dulbecco’s Modified Eagle Medium DNA Deoxyribonucleic Acid DP Dispersed Phase ECM Extracellular Matrix ESI-MS Electrospray Ionization Mass Spectroscopy FBS Fetal Bovine Serum FOV Field of View GDA Gastroduodenal Artery Gy Gray HCC Hepatocellular Carcinoma xiv  HEC Hydroxyethyl Cellulose HU Hounsfield Units I-XCA Iodine-Based X-Ray Contrast Agent IARC International Agency of Research on Cancer ICP-MS Inductively Coupled Plasma Mass Spectroscopy ID Inner Diameter IMDM Iscove’s Modified Dulbecco Medium IMRT Intensity Modulated Radiotherapy IR Infrared keV Kiloelectron Volt kVp Kilovoltage Peak LD50 Median Lethal Dose LOI Line-of-Interest mA Milliampere MAA  Macroaggregated Albumin MALDI MS Matrix Assisted Laser Desorption Mass Spectroscopy mAs Milliampere-Second MBF Modified Barrier Facility 𝑀𝑛 Number Average Molecular Weight MRI Magnetic Resonance Imaging ms Millisecond MS Microspheres  𝑀𝑤 Weight Average Molecular Weight NF Nanofiber NIH National Institutes of Health NIST National Institute of Standards and Technology NMR Nuclear Magnetic Resonance 𝑂𝐷 Optical Density  OD Outer Diameter PCL Polycaprolactone 𝑃𝐷𝐿 Population Doubling Level  xv  Pen-Strep Penicillin-Streptomycin PET Positron Emission Tomography PHA Proper Hepatic Artery PLA Poly(𝐿-Lactic Acid) PMMA Poly(Methyl Methacrylate)  PTCA Percutaneous Transluminal Coronary Angioplasty PVA Poly(Vinyl Alcohol) 𝑄𝐶𝑃 Continuous Phase Flow Rate 𝑄𝐷𝑃 Dispersed Phase Flow Rate 𝑅2 Coefficient of Determination Re-MS Microspheres Doped with Rhenium (i.e., Non-Radioactive) Re-NPs Rhenium Nanoparticles ROI Region of Interest SARRP Small Animal Radiation Research Platform SEM Scanning Electron Microscopy SPECT Single Photon Computed Tomography SQ Subcutaneous Injection TAE Transarterial Embolization 𝑇𝑔 Glass Transition Temperature TLC Thin Layer Chromatography 𝑇𝑚 Melting Temperature TOF MS Time-of-Flight Mass Spectroscopy UBC University of British Columbia UV/Vis Ultraviolet/Visible  VGH Vancouver General Hospital VOI Volume-of-Interest XCA X-Ray Contrast Agent 𝑍 Atomic Number 𝜇CT Micro-Computed Tomography 𝜇𝑋  Linear X-Ray Attenuation Coefficient 𝜇𝑋/𝜌 Mass X-Ray Attenuation Coefficient xvi  Acknowledgements  The past few months have been a roller coaster of emotions. I finished my experiments, wrote my dissertation, and started to prepare for the next chapter in my career. This dissertation is the culmination of my journey through a Ph.D. which has been like climbing a high mountain one step at a time accompanied by hard work, enthusiasm, curiosity, hope, and sometimes, frustration. This has truly been a self-discovery experience and I am excited to see how it will unfold in the next few weeks. Now that I am approaching the summit, I want to take a moment to acknowledge everyone who helped me to accomplish this huge task.    Dr. Urs Hafeli, I am greatly indebted to you. You took me into your laboratory when I had hardly any experience and shaped me into the scientist I am today. Thank you for your confidence in me and my research and for your advice, support, and guidance throughout all these years. I feel very fortunate to have had the opportunity to work with you. I admire you.  Dr. Katayoun Saatchi, you were an important pillar of my Ph.D. Thank you for doing everything in your power to help me succeed in my experiments and make me feel confident about my work. I am forever in debt.  Thank you for joining me in this journey, Drs. Marcel Bally, Anna Celler, Adam Frankel, and Peter Soja. You helped me to take my research in exciting directions with your knowledge and experience. I am very grateful for your words of encouragement.  Drs. David Cormode, Cheryl Duzenli, and Michael Wolf, thank you for accepting the invitation to be part of my final oral examination. I will always remember that day with joy.  Dr. Pedro Esquinas, I am very lucky to have you as my friend. You were always there when I needed help and contributed enormously to my research. Thank you very much.  Dr. Cristina Rodríguez, I appreciate your expert advice. Thank you for caring so much about me and my research and for keeping me company at CCM.  Jovan Gill and Selin Jessa, you taught me a lot. Thank you for all your hard work Thank you for sharing your expertise with me, Drs. Bradford Gill, David Liu, Igor Moskalev, and Yogesh Thakor. You helped me to take my research one step further. I owe you. John Schipilow, I really enjoyed our scientific conversations. Thank you for going above and beyond to help me. Dr. Laura Mowbray and Rhonda Hildebrandt, thank you for all your support. You were always friendly faces at CCM.  xvii  I acknowledge the National Council of Science and Technology of Mexico (CONACYT) for financial support. My Ph.D. was possible due to the scholarship that was awarded to me.  These past five years have been full of challenges, with many ups and downs. To all of my friends, thank you for keeping me smiling and feeling alive even in my toughest days. There is not enough space here to thank you all, but I promise I will speak to each of you in person.  Jennifer Brown, have you considered a career in editing? Thank you for always listening.     Zeynab Nosrati, it was a pleasure working with you. I admire your calmness.  Arnab Ray, thank you for sharing your creative ideas with me. Lennart Bohrmann and Dana Lambert, I am glad we shared an office. Thank you for never judging me for eating excessive amounts of chocolate when I was writing my dissertation.  Dr. Irene Andreu, Marysol García, Vongai Nyamandi, and Julia Varela, I feel lucky we crossed paths in this experience. Thank you for all the good times.  Dr. Miguel Suástegui, your messages kept me sane. I miss you.  Alejandro Esquivel, Hugo Morales, and Sandra Soto, you are forever part of my family. Josh Belford, José De Anda, Taylor Neumann, Trish Rivers, Ondrea Ross, Jonathan Steele, and Sebastien Trudeau, you are the best. Thank you for reminding me of the little things in life.  On a very personal note, I want to thank my parents, José Carlos and Diana, for believing in me and encouraging me to follow my dreams. Thank you for the selfless love, care, pain, and sacrifice you did to shape my life. You are my biggest inspiration. Daniela, your strength makes me stronger. You are my idol. José Alberto, thank you for always having my back. You are my big brother (not the other way around). To all of you, thank you for standing by my side and trusting every one of my choices. I miss you every day. You are always in my heart and on my mind. This dissertation, and everything I do, is for you.    I want to thank as well a very special person, my partner in crime, best friend, and endless source of chocolate, Brent. Thank you for your continued and unfailing love, support, and understanding during the most challenging times of my Ph.D. You were always around at times I thought it was impossible to continue and helped me to keep things in perspective. You believed in me when I doubted myself and gave me the courage to stay strong. Thank you for putting up with me. I would not have been able to get here without you. This dissertation is also for you.  I am so fortunate to have all of you in my life. Thank you.   José Carlos De La Vega  xviii  Dedication                  To very special people in my life: my parents, José Carlos and Diana; my sister, Daniela;  my brother, José Alberto; Thank you for your unconditional love  and unwavering support  1  Chapter 1 Introduction  1.1 Overview of the Dissertation  The research described on this dissertation centers around rhenium (Re), a transition metal with atomic number (𝑍) of 75. Since its discovery in 1925, rhenium has been recognized as a chemically versatile element with rich coordination chemistry. A wide range of thermodynamically stable non-radioactive and radioactive coordination compounds of rhenium have been synthesized and characterized. The utilization of some of these rhenium complexes in nuclear medicine has already been investigated, particularly of complexes of 188Re (a mixed 𝛽− and 𝛾-emitting radioisotope of rhenium). There might be many more applications in medical imaging for rhenium complexes and this dissertation aims to establish a few more.  The development of contrast agents for X-ray imaging, or X-ray contrast agents (XCAs), is a field of research that has recently garnered a lot of attention. Before radiographic examinations, XCAs are administered to patients to improve visualization of internal body structures, which, with the exception of bone, lack sufficient contrast. Most XCAs are currently made of iodine (I, 𝑍 = 53). However, X-ray attenuation can be further increased with the use high-𝑍 elements. The X-ray attenuation is dependent on the atomic number and the density. Thus, high-𝑍 elements have extensively been investigated in the search of new XCAs. The majority of studies have focused on the use of gold (Au, 𝑍 = 79), but the following elements have also been examined: bismuth (Bi, 𝑍 = 83), tantalum (Ta, 𝑍 = 73), ytterbium (Yb, 𝑍 = 70), and gadolinium (Gd, 𝑍 = 64). Despite its high atomic number, little is known about the potential utilization of rhenium in X-ray imaging.  In this dissertation, the X-ray attenuation of rhenium was thoroughly investigated. Upon demonstrating the usefulness of rhenium in X-ray imaging, the following medical devices were produced using biodegradable and biocompatible polymers doped with non-radioactive rhenium complexes: 1) a radiopaque electrospun scaffold and 2) a sample of radiopaque uniformly-sized microspheres (MS). As a result of the success with the preparation of size-defined MS with a narrow size distribution, radioactive MS labeled with a 188Re complex were also produced. These MS can be imaged by single photon emission computed tomography (SPECT), a medical imaging modality which detects 𝛾 photons. The 𝛽− particles emitted by 188Re have a high energy 2  and thus cause damage to the deoxyribonucleic acid (DNA) when they are absorbed by the cells, resulting in programmed cell death or apoptosis. In other words, they have a high cell killing power, yielding these “imageable” MS theranostic (i.e., simultaneously diagnostic and therapeutic), with quantifiable radiotherapeutic potential in cancer. Specific applications were proposed for all of these newly developed medical devices. The general objective of this dissertation is to provide the framework for the incorporation of non-radioactive and radioactive rhenium complexes into polymer-based devices for medical imaging. The foundations of this work rely on the following three aims:   Aim 1  To evaluate the potential use of rhenium in X-ray imaging by assessing image quality and absorbed dose.   Aim 2  To incorporate a radiopaque component onto otherwise radiolucent catheters by coating them with an electrospun scaffold doped with a rhenium complex.   Aim 3  To prepare, characterize, and image uniformly-sized MS that 1) are doped with a rhenium complex (i.e., Re-MS) and 2) are labeled with a 188Re complex (i.e., 188Re-MS).  Herein, these aims are addressed in the following research chapters:  Aim 1      Chapter 2: Investigation of the X-Ray Attenuation of Rhenium.  Aim 2      Chapter 3: Development and Imaging of Radiopaque Electrospun Scaffolds.  Aim 3      Chapter 4: Development and Imaging of Radiopaque and Radioactive Microspheres.  The rest of this introductory chapter provides some general background on the topics which are common across the three research chapters, specifically the relevant chemical and radioactive properties of rhenium and the state of the art in the development of XCAs. The necessary background related to each of the aforementioned specific aims and a detailed description of the methods is then given in the respective research chapters. The concluding chapter offers an overall analysis of the applications of the work described in the research 3  chapters, focusing on this research’s significance and contribution, strengths and limitations, and future directions.   1.2 Chemical and Radioactive Properties of Rhenium  Rhenium is a high-𝑍 silvery-white transition metal with atomic mass of 186.2 Da. In the periodic table, it is located in the third-row of Group VII. Rhenium occurs naturally as a mixture of two isotopes: 185Re (37.4%), which is stable, and 187Re (62.6%), which is unstable but has a very long half-life (41.2x109 years). More than 25 radioisotopes of rhenium have been identified [1-3]. Among all of them, 188Re is of special interest in nuclear medicine.  188Re decays to the stable osmium isotope 188Os with a half-life of 17.0 h by emission of 𝛽− particles with a maximum energy of 2.12 MeV, making 188Re-labeled radiopharmaceuticals suitable for radiotherapy. These 𝛽− emissions have a relatively small penetration distance in tissue: 11.0 mm (maximum distance) and 3.8 mm (mean distance). At the same time, 155 keV 𝛾 photons are emitted with an abundance of 15.6%, allowing for in vivo imaging of the biodistribution of 188Re-labeled radiopharmaceuticals by SPECT for utilization in personalized dosimetry calculations (Table 1.1) [4]. A convenient and cost-effective method to produce 188Re is in a generator by the decay of 188W, a radioisotope of tungsten with a half-life of 69.4 days. The 188W/188Re generator is a long-term (at least 4 to 6 months) continuous source of no-carrier-added 188Re (i.e., essentially free from stable isotopes of rhenium) [5-7].  Table 1.1. Energy of Most Abundant 188Re Emissions. 188Re is a mixed 𝛽− and 𝛾 emitter and can thus be used for therapy and diagnosis, making it a true theranostic radioisotope.   Type Energy (keV) Yield (%) 𝛽− 2,120 (Maximum) 760 (Mean) 100 𝛾 155 15.6 478 1.1 633 1.4 829 0.4 X-Ray (k𝛼) 61 – 63  3.8 X-Ray (k𝛽) 71 – 73 1   Interestingly, rhenium and technetium are congeners with similar chemical properties [8,9]. Many well-characterized radioactive complexes of technetium, specifically 99mTc complexes, are presently available because of the popularity of 99mTc-labeled radiopharmaceuticals in medical 4  imaging [10-15]. Due to the similarities in size and geometry of complexes of these two elements, the same ligands used to bind technetium can be used to bind rhenium [16-19]. This has ultimately led to the synthesis of a myriad of non-radioactive and radioactive rhenium complexes [20-25]. In this dissertation, [ReOCl(MePO)2] (a non-radioactive rhenium phosphinophenolate complex) will be utilized in the study described in Chapter 3, whereas [Re(CO)3]+ and [188Re(CO)3]+ (non-radioactive and radioactive rhenium(I) tricarbonyl complexes) will be utilized in the studies described in Chapter 4.   Rhenium is a chemically versatile element which can exist in oxidation states from -1 to +7 [26]. Compared with their technetium analogues, rhenium complexes are thermodynamically more stable in their higher oxidation states. For instance, rhenium complexes oxidize in vivo to perrhenate (ReO4-). This transformation occurs because perrhenate features rhenium in the oxidation state of +7, which is rhenium’s preferred oxidation state [8,26]. This is quite advantageous for biological applications because the pharmacokinetics of perrhenate is well-understood: it is taken up avidly by the thyroid, retained there with a biological half-life of 12 h, and finally excreted via the kidneys [27-29]. This biodistribution profile has also been reported by Häfeli et al. after making 186/188ReO4- and injecting it into mice [30].   1.3 Medical Imaging and X-Ray Contrast Agents  Briefly, medical imaging aims to create images of the internal structures of the body to diagnose diseases and monitor response to treatment. Among all different imaging modalities available, X-ray imaging is by far the most commonly utilized because it has a good spatial resolution (50 to 200 𝜇m) and is one of the fastest imaging procedures to complete (around 1 s for almost any modern equipment) [31]. X-ray imaging is based on the principle that photons are attenuated when they pass through the body. The X-ray attenuation is measured by a detector to create two-dimensional (2D) and/or three-dimensional (3D) images. The most popular X-ray imaging technique is planar X-ray imaging (or projection radiography), where 2D images are acquired on film. Currently, the term digital radiography is preferred when photons are detected electronically [32]. Fluoroscopy is another extensively used technique, especially in X-ray imaging-guided catheterizations because it produces real-time 2D images. First introduced in the 1970s, computed tomography (CT) plays a mainstay role in the staging and imaging-guided intervention of various diseases. In a common CT configuration, the 5  X-ray tube and the detector (specifically, an array of detectors) rotate around the patient in a synchronized fashion in order to build a 360°dataset. The gantry is the “donut”-shaped part of the scanner which houses the X-ray tube and the detector, positioned opposite to each other. Most modern scanners have as many as 320 rows of detectors and operate in a helical fashion. This means that the X-ray tube and the detector continuously rotate while the patient is constantly moved out of the scanner. Through mathematical algorithms, the dataset is reconstructed as a 3D image [33,34].  CT has garnered much attention due to the development of micro-CT (𝜇CT), which allows the assessment of microstructures with high spatial resolution (<50 𝜇m). This technique has been particularly useful in the analysis of bone and vascular structures in both research and preclinical studies [35-37]. Another variation of CT popular in implant dentistry is cone beam CT (CBCT), which is simply a compact and faster version of CT. The time needed to complete a scan is reduced in CBCT through the utilization of a divergent, cone-shaped X-ray beam instead of a narrow, fan-shaped X-ray beam, used in CT [38-41].  All X-ray imaging techniques require the use of XCAs to enhance visualization of tissues and organs. The majority of XCAs are solutions of iodinated compounds, which are administered to patients before radiographic examinations via intravenous or intra-arterial injections. Some XCA used for imaging of the gastrointestinal tract are, however, prepared with barium compounds (specifically barium sulfate, or BaSO4) [42]. The evolution in the structure of iodine-based XCAs (I-XCAs) moved from inorganic iodine (specifically sodium iodide, or NaI) to organic mono-, di-, and tri-iodinated compounds, from lipophilic to hydrophilic compounds, from ionic to non-ionic compounds, and more recently, from monomers to dimers (Figure 1.1). These modifications aimed to: 1) increase the number of atoms of iodine in the compounds and 2) reduce the incidence of adverse effects associated with the use of I-XCAs.      6    Figure 1.1. Iodine-Based X-Ray Contrast Agents. The chemical structures of representative I-XCAs are shown: uroselectan A (an ionic mono-iodinated monomer), uroselectan B (an ionic bi-iodinated monomer), diatrizoate (an ionic tri-iodinated monomer), iohexol (a non-ionic tri-iodinated monomer), iopromide (a non-ionic tri-iodinated monomer), iopamidol (a non-ionic tri-iodinated monomer), and iotrolan (a non-ionic hexa-iodinated dimer).   A clinical issue with I-XCAs is that they are rapidly excreted by the kidneys, requiring the administration of repeated large doses to achieve relatively good contrast. While the dose of iodine is usually only up to ~3 g in lumbar, thoracic, cervical, and columnar radiographic examinations [43,44], it could be as high as ~30 g in coronary CT angiography [45] and ~100 g in angiography and angioplasty [46]. Moreover, I-XCAs have been linked to a number of adverse effects with an estimated incidence between 1 and 12% [47]. Adverse effects can range from mild reactions, such as itching and emesis, to even life-threating emergencies. Common severe adverse effects include hypersensitivity reactions, thyroid dysfunction, anaphylaxis, and nephropathy (sometimes referred to as contrast media-induced nephropathy, a common condition in patients with pre-existing renal impairment) [48-50].  In addition to the well-reported toxicity of I-XCAs, iodine attenuates high-kVp X-rays (>80 kVp) less than other elements and, as a consequence, images exhibit less contrast and are noisier [51,52]. The X-ray tube potential is the voltage applied to the X-ray tube in radiographic examinations, reported in units of kVp (i.e., kilovoltage peak). Most scans are performed between 24 (e.g., mammography) and 140 kVp (e.g., chest CT) [53]. The suboptimal X-ray attenuation of iodine is particularly an issue when imaging average-sized and large patients, which is typically done at 100 to 140 kVp [54-56].  Considering the above-mentioned problems, an active field of research is the development of iodine-free XCAs. As a rule, materials with greater atomic number and density exhibit a higher X-ray attenuation. More precisely, the X-ray attenuation is dependent upon the atomic 7  number raised to the third power [57]. Hence, research has predominantly focused on replacing iodine (𝑍 = 53) with elements with a greater atomic number. Over the last few years, several nanoparticles (NPs) loaded with high-𝑍 elements have been developed for this application which have two additional advantages over I-XCAs: they have a higher circulation half-life and easily contain a larger number of radiopaque atoms [58].   1.3.1 Development of X-Ray Contrast Agents with High-𝒁 Elements  Among all the elements with an atomic number greater than iodine, gold (𝑍 = 79) took the leading role in the development of iodine-free NP-based XCAs. One of the advantages of gold NPs (i.e., AuNPs) is that they have been investigated for many years for other biomedical applications (e.g., photoacoustic imaging, photothermal ablation, and DNA detection) [33]. Historically, gold salts were first utilized in the 1940s for the treatment of rheumatoid arthritis, showing minimal toxicity [59]. More recently, AuNPs have been shown to exhibit good biotolerability (Median Lethal Dose, or LD50 = 3.2 g of gold per kg) [60-62]. Through both phantom and animal studies, many research groups have examined the X-ray attenuation of AuNPs of various sizes, between 2 and 20 nm. These studies have shown that gold exhibits a better X-ray attenuation than iodine at the same concentration [62-65]. Not only perfectly spherical AuNPs have been developed, but also a wide number of gold-containing nanostructures with different shapes, including nanorods, nanoclusters, nanoshells, nanocages, nanocubes, and nanoprisms. For some of these nanostructures, animal studies have been conducted to appraise their biodistribution, showing an increased circulation half-life compared to I-XCAs (and thus, prolonged contrast enhancement) [66-68]. A limitation of gold is that it is more expensive than iodine, which might hinder the utilization of AuNPs in clinical practice. Therefore, other elements, like bismuth (𝑍 = 83), have also been explored. Bismuth has the advantage of decomposing in vivo to small and renally clearable Bi(III) species as a result of its tendency to oxidize and hydrolyze in water (i.e., it does not accumulate in the body) [69]. Examples of bismuth NPs (i.e., BiNPs) under investigation are: ultra-high payload BiNPs [70], bismuth glyconanoparticles (BiGNPs) [71] and bismuth(III) sulfide (Bi2S3) NPs [72,73]. Tantalum (𝑍 = 73) is another element which has garnered attention as an XCA, especially upon reaction with oxygen to form tantalum pentoxide (or Ta2O5). The use of tantalum pentoxide as an XCA was actually reported many years ago in bronchography, 8  after delivery as a powder aerosol via the trachea [74]. Some of the advantages of this compound are its chemical stability, high solubility in water, and biocompatibility. Current efforts revolve around the production of NPs comprised of a core made of tantalum pentoxide [75,76].  Elements of the lanthanide series, such as ytterbium (𝑍 = 70) and gadolinium (𝑍 = 64), have also been explored in the development of NP-based XCAs [77]. Interestingly, lanthanides are excellent contrast agents for magnetic resonance imaging (MRI) [78,79]. Therefore, an ongoing field of research is the production of NPs loaded with a mixture of lanthanides for multimodal imaging: that is, contrast agents for both X-ray imaging and MRI. In this regard, NPs containing mixtures of ytterbium/erbium [80] and ytterbium/thulium [81] are under development.  While much effort is ongoing to develop iodine-free XCAs, mainly I-XCAs are routinely employed in medical procedures. A few cases will be explored in this dissertation. For example, I-XCAs are commonly administered in X-ray imaging-guided catheterizations. Moving catheters inside the body requires knowledge of the exact location and orientation of their distal tip at all times. Since most catheters are radiolucent (i.e., transparent to X-rays), I-XCAs are used to help physicians advance the catheters through the blood vessels. In Chapter 3, reducing the radiolucency of catheters with rhenium will be studied. Another example is in embolization therapy, or embolotherapy, where MS are administered through a catheter to obstruct a blood vessel. Embolic MS have been found to be useful to control and/or prevent bleeding, eliminate abnormal connections between arteries and veins, treat aneurysms, and even block the blood supply to tumors. Since MS are naturally radiolucent, they are blended prior to administration with I-XCAs, which act as a surrogate of the location of the MS in the body. Unfortunately, visibility diminishes soon after injection due to XCA washout, leaving the final location of the MS unknown. This case will be explored in Chapter 4.  Despite the high atomic number of rhenium (𝑍 = 75), its potential utilization in X-ray imaging has only been reported in a few recent manuscripts by Krasilnikova et al. [82-85]. Interestingly, the effect of the X-ray beam’s mean energy on the X-ray attenuation of rhenium has not been studied in depth. This is precisely the aim of the following chapter, which describes a study conducted to compare the X-ray attenuation of rhenium and iodine over a wide range of experiment conditions.   9  Chapter 2 Investigation of the X-Ray Attenuation of Rhenium  The objective of this chapter is to provide evidence of the potential utilization of rhenium in X-ray imaging. To accomplish this, a study was conducted to compare the X-ray attenuation of solutions containing rhenium and iodine. The comparison focused on the evaluation of the image quality, which was assessed experimentally using preclinical and clinical equipment, and the absorbed dose, which was estimated for a clinical equipment using Monte Carlo.     2.1 Background  Since their discovery in 1895 by Röntgen, the nature of X-rays has been extensively investigated and many of their properties have been unraveled, making possible their utilization in medical applications. X-rays are a form of electromagnetic radiation and thus exhibit wave-like and particle-like properties, whereby interactions are collisional in nature [86]. Although originally named to underline the fact that the origin of their energy was unknown, our understanding of the mechanisms of the production of X-rays and their interaction with matter has been greatly enhanced over the last century [87]. These mechanisms are in explained in the next section. Furthermore, important parameters to describe image quality and absorbed dose in X-ray imaging are defined.   2.1.1 X-Ray Production and Interaction with Matter  Modern X-ray instruments have a very complex circuit, but they all rely on an X-ray tube that operates under the same principle as the one that ultimately led to Röntgen’s discovery. Figure 2.1 shows a conventional X-ray tube, which basically consists of a glass enclosure that has been evacuated to high vacuum. Situated 1 to 2 cm apart from each other, two electrodes, the cathode and the anode, are found inside the X-ray tube. The cathode consists of a filament made of tungsten which emits electrons when it is heated. The anode consists of a thick copper rod with a target located at the edge (usually a small piece of tungsten). An X-ray generator provides the source of electrical voltage to energize the X-ray tube. This is accomplished by connecting high-voltage cables to the electrodes. The circuit is completed by connecting the tungsten filament with a low-voltage power source using a separate, isolated cable [57].  10    Figure 2.1. X-Ray Tube. The main elements of the X-ray tube are the electrodes, located inside the enclosure made of glass. Other representative parts are the cathode’s filament, which is made of tungsten, and the anode’s rod and target, which are made of copper and tungsten, respectively. A small window made of beryllium is placed on the trajectory of the X-ray beam to maximize flux transmission.   A sequence of specific events is required in order to produce X-rays. First, an X-ray tube potential is applied between the cathode and the anode (between 24 to 140 kVp in most radiographic examinations [53]). Upon activation, electrons are emitted from the filament and accelerated towards the anode, achieving high velocities before striking the target. Next, the electrons interact with the target by one of two different mechanisms that lead to the propagation of electromagnetic radiation in the form of X-rays. These mechanisms are: Bremsstrahlung emissions (from the German “bremsen”, which means “to brake”, and “Strahlung”, which means “radiation”; i.e., braking radiation) and characteristic X-ray radiation. Finally, photons emerge through a thin window usually made of beryllium, located in the X-ray tube to maximize flux transmission [88]. The Bremsstrahlung emissions are generated by the sudden loss of velocity of the high-speed electrons at the target. In general, when electrons pass near the nuclei of the atoms of the target, they are deflected from their path by the action of Coulomb forces. This phenomenon results in partial or complete loss of the electrons’ kinetic energy, which is then converted into photons. Since the electrons may undergo one or more Bremsstrahlung interactions and these interactions may result in differential losses of kinetic energy, the photons might have any energy up to the initial energy of the electrons, which corresponds to the X-ray tube potential supplied [57,88].  The characteristic X-ray radiation is produced when electrons interact with the atoms of the target by ejecting an electron from one of the atom’s inner shells. This process creates a vacancy in the shell, which is then filled by an electron from one of the outer shells. The energy released 11  when one of the electrons falls down to fill the vacancy is converted into photons. Contrary to Bremsstrahlung interactions, characteristic X-ray radiation is thus released at a discrete energy which is characteristic of the atom in question [57,88].   The energy of each photon ℎ𝜐 is typically expressed in keV (i.e., kiloelectron volt; 1 keV = 1.6x10-16 J), where ℎ is the Planck’s constant (6.6x10-34 J s) and 𝜐 is the photon’s frequency. The photons transfer their energy to the medium (e.g., tissues and organs) by interacting with electrons which then transfer energy to the medium through atomic excitations and ionizations. The interactions between photons and electrons occur through various mechanisms, such as: photoelectric effect, Compton effect, Rayleigh scattering, and pair production. The greater contribution to the attenuation of the X-ray beam comes from the photoelectric effect and the Compton effect [89]. These two phenomena are illustrated in Figure 2.2.   The photoelectric effect refers to the interaction between an incident photon and an inner shell electron which has a binding energy less than the energy of the incident photon (i.e., ℎ𝜐 > 𝐸𝐵, where 𝐸𝐵 is the electron shell binding energy). During this process, the entire energy of the incident photon is transferred to the electron, which results in the ejection of the electron from its shell. The energy of the ejected electron, known as photoelectron, is equal to: ℎ𝜐 – 𝐸𝐵. The photoelectric effect usually involves an electron of the K shell, which is the closest shell to the nucleus. The vacated shell is subsequently filled by an electron from an upper shell, such as the L shell. This causes the release of characteristic X-ray radiation with energy equal to the difference in binding energies between the electron shell with the initial vacancy (i.e., the “acceptor” electron shell) and the electron shell with the final vacancy (i.e., the “donor” electron shell) [89,90].  In the Compton effect, the energy of the incident photon is much greater than the binding energy of the electron (i.e., ℎ𝜐 ≫ 𝐸𝐵). During this process, the incident photon transfers a fraction of its energy to the electron and is scattered at an angle 𝜃. The electron, on the other hand, is scattered at an angle 𝜑. For illustrative purposes, it is useful to imagine the electron at the origin of a coordinate system where the incident photon is directed in the positive direction along the 𝑥-axis. The scattered photon can travel in any direction, with 𝜃 taking values between 0 and 180° relative to the 𝑥-axis (i.e., the trajectory of the incident photon). The scattered electron, known as recoil electron, can only travel forward, with 𝜑 adopting values between 0 and 90° relative to the to the 𝑥-axis [89,90].  12    Figure 2.2. Photoelectric Effect and Compton Effect. The cartoon depicts an illustrative summary of the most common types of X-ray interactions with matter. The photoelectric effect results in the complete transfer of the energy of the photon to the electron (or photoelectron). Conversely, in the Compton effect, the transfer of energy is shared between the photon, scattered at an angle 𝜃, and the electron, scattered at an angle 𝜑 and specifically known as recoil electron. Furthermore, some photons may traverse the medium without any interaction.  The aforementioned mechanisms reduce the intensity of the incident X-ray beam by removing photons via absorption or scattering events. The attenuation of the X-ray beam is schematically depicted in Figure 2.3. The following equation shows that there is an exponential relationship between the intensities of the incident and the transmitted X-ray beams passing through a material of thickness 𝑥.   𝐼(𝑥) = 𝐼𝑜𝑒−𝜇𝑋𝑥          (Equation 2.1)  where 𝜇𝑋 is the linear X-ray attenuation coefficient. The thickness of the material is usually given in cm, so the linear X-ray attenuation coefficient, which represents the probability of attenuation per centimeter of the material, has units of cm-1. The linear X-ray attenuation coefficient depends on the energy of the photons as well as on the atomic number and the density of the attenuator. The differences in the contrast of images acquired in radiographic examinations are caused by the differences in the linear X-ray attenuation coefficient of the tissues being scanned. In other words, X-ray images are “maps” of the spatially varying linear X-ray attenuation coefficient [90].  13    Figure 2.3. X-Ray Beam Attenuation. (A) Schematic of the attenuation of an X-ray beam after exiting the X-ray tube. The number of photons in the incident X-ray beam is reduced after passing through the attenuator via absorption or scattering events. A collimator is found in most X-ray equipment so that only photons traveling parallel to a specified direction are allowed through. (B) The attenuation of the incident X-ray beam occurs exponentially according to the equation: 𝐼(𝑥) = 𝐼𝑜𝑒−𝜇𝑋𝑥, where 𝜇𝑋 is the linear X-ray attenuation coefficient..   To suit the needs of specific applications, the incident X-ray beam can be modified by varying the following user-selectable parameters: X-ray tube potential, current, exposure time, and amount of filtration (i.e., thicknesses of the filters).  As the X-ray tube potential is increased, both the photon count and the photon energy are increased [88]. This is exemplified graphically in Figure 2.4, which depicts the X-ray spectra at 50 and 120 kVp in the presence of a 0.6 mm thick copper sheet. Basically, when an X-ray tube potential of 120 kVp is supplied to the X-ray tube, photons with a continuous distribution of energies up to 120 keV are produced. At 50 kVp, however, only photons with energies up to 50 keV are generated.    14   Figure 2.4. Effect of the X-Ray Tube Potential on the X-Ray Spectra. For a specific X-ray beam, the X-ray spectrum can be shown as a graph where the photon count is on the 𝑦-axis and the photon energy is on the 𝑥-axis. The mean energy of the X-ray beam occurs at the peak maximum of the curve. The maximum energy which the photons can potentially have is equal to the X-ray tube potential. The low photon count below 20 keV is due to filtration of the X-ray beam with a 0.6 mm thick copper sheet.   The current, in mA (i.e., milliampere), and the exposure time, in ms (i.e., millisecond) also have an effect on the photon count, but they do not change the photon energy. The current is defined as the number of electrons traveling between the electrodes as a result of an applied X-ray tube potential. Thus, it determines the number of photons produced at a fixed exposure time. The exposure time is the duration of each irradiation, generally always kept as short as possible. These two parameters are oftentimes given simply as the current ∙ exposure time product, in mAs (i.e., milliampere-second) [88]. The sources of filtration reduce the photon count and increase the mean energy of the X-ray beam by removing low-keV photons. However, they do not change the maximum energy of the X-ray beam, which is specific of each X-ray tube potential [91,92]. In other words, as the sources of filtration increase, the X-ray beam becomes proportionally richer in high-keV photons. This phenomenon is known as beam hardening and it increases the penetrability of the X-ray beam (i.e., the distance the photons travel in matter) [93,94]. The sources of filtration can be classified as either inherent or additional. The inherent filtration consists of all the materials that the photons encounter as they exit the X-ray tube, such as the X-ray window made of beryllium. Also, aluminum (Al, 𝑍 = 13) sheets with thicknesses up to 1 cm are usually found in most instruments. The additional filtration includes all the materials which are placed between the X-ray tube and the detector, such as aluminum and copper (Cu, 𝑍 15  = 29) sheets. These materials are placed in the direct path of the X-ray beam, close to the X-ray tube. The body also acts as a source of additional filtration [91,92].  We conducted a study where rhenium- and iodine-based samples were scanned at a wide range of X-ray tube potentials, between 50 and 220 kVp, with increasing thicknesses of copper sheets, between 0 and 1.6 mm. The current and the exposure time were chosen to avoid saturation of the detector with photons. If the detector is saturated, extremely bright images are produced. Nonetheless, if not enough photons reach the detector, then image noise is increased. For each scan, the quality of the images was evaluated to compare the X-ray attenuation of the samples. Usually, there is a trade-off between image quality and radiation dose. In the clinical setting, therefore, the optimization of the settings of radiographic examinations also requires considerations of radiation dose [95,96]. In this study, we estimated the radiation dose in a few clinical relevant scans (in terms of X-ray tube potential and amount of additional filtration). Some important terminology regarding image quality and radiation dose is provided in the next section.  2.1.2 Assessment of Image Quality and Radiation Dose  As explained in the previous section, images are created by the detection of the transmitted X-ray beam. Radiopacity, or radiodensity, is the ability of materials of inhibiting the passage of photons. Radiolucency, on the other hand, is the inability of materials of obstructing the passage of photons (i.e., it refers to materials which are transparent to X-rays). On images, radiopaque materials (e.g., bone) appear white, whereas radiolucent materials (e.g., air) appear black. The images are thus made of pixels with various shades of grey. The tonal value of each pixel, referred to as pixel intensity, is associated with a grey value [97].  The quality of the images is quantified in terms of their contrast and noise. Contrast is the result of the differential X-ray attenuation of tissues and organs. Noise refers to the random fluctuations across the images that affect the detection of structures with low contrast [95]. A parameter of image quality which takes into account these two definitions is the contrast-to-noise ratio (𝐶𝑁𝑅). The contrast is calculated as the difference between the mean pixel intensity of a material of interest and the mean pixel intensity of the background. The noise is equal to the standard deviation of the mean pixel intensity of the background [98]. These two concepts are exemplified graphically in Figure 2.5. A useful image requires structures of interest to be 16  distinguished against the background. The 𝐶𝑁𝑅 is typically calculated considering water or air as the background, but water is preferred because it has a similar X-ray attenuation than soft tissue [99-101]. For our study, the 𝐶𝑁𝑅s of our rhenium- and iodine-based samples were calculated relative to water using Equation 2.2.     Figure 2.5. Contrast and Noise. The graph shows the variation in the grey values of the pixels across a material of interest surrounded by background. Generally, high noise may still allow differentiation of various structures within a single image provided the contrast is significantly greater than the noise.  𝐶𝑁𝑅 = 𝜄?̅?− 𝜄?̅?2𝑂 𝜎𝐻2𝑂          (Equation 2.2)  where 𝜄 ̅is the mean pixel intensity, 𝜎 is the standard deviation of the mean pixel intensity, and 𝑖 represents a material of interest. The same definitions apply for 3D images, which are made of voxels. The only difference is that the term pixel intensity is replaced with voxel intensity.  The Hounsfield scale is also commonly utilized to quantify the contrast, particularly in CT. On this scale, the attenuation of a material of interest is reported in Hounsfield units (HU). The Hounsfield scale is a linear transformation of the linear X-ray attenuation coefficient that takes as a reference the attenuation of air and water (Equation 2.3).  Attenuation = 1,000 𝜇𝑋𝑖− 𝜇𝑋𝐻2𝑂𝜇𝑋𝐻2𝑂− 𝜇𝑋𝐴𝑖𝑟      (Equation 2.3)  where 𝜇𝑋 is the linear X-ray attenuation coefficient and 𝑖 represents a material of interest. The Hounsfield scale is set by defining the attenuation of water as 0 HU and the attenuation of air as -17  1,000 HU. In general, soft tissues range between -100 and 100 HU and bones range between 400 and 1,000 HU [102].  The quality of an image and the anatomical detail seen within it depend on the characteristics of the equipment and the X-ray tube potential applied (and thus, the photon energy). The equipment, for instance, has an effect on the resolution, or sharpness, of the image (i.e., the ability to differentiate between two structures located at a small distance apart). In general, the use of a higher X-ray tube potential will improve the quality of the image within certain limits, but will give the patient a higher radiation dose [95,96]. Hence, it is important to recognize the level of image quality that is required to make an accurate diagnosis and determine the settings that provide that level of image quality with the minimum radiation dose.  The first step in the absorption of X-rays is the production of photoelectrons (via the photoelectric effect) and recoil electrons (via the Compton effect) Therefore, X-rays are indirectly ionizing radiation. In other words, they do not produce biological damage themselves, but when photons are absorbed in the material through which the pass (e.g., tissues and organs), they give up their energy to produce fast-moving charged particles. The absorbed dose is defined as the mean energy taken up by tissues and organs per unit mass, typically reported in Gy (i.e., gray) [103].  Another quantity of radiation dose is the entrance surface dose, which is the dose at the point where the X-ray beam enters the body. The entrance surface dose, which is estimated through computational modeling (e.g., Monte Carlo [104]), takes into consideration the air-kerma and the backscattered radiation [105]. The air-kerma is the mean energy deposited in air per unit mass, also reported in Gy. Generally, the air-kerma is considered a good indicator of radiation risk when different settings and techniques are compared because it can be easily measured with an ionization chamber [106,107].   2.2 Materials and Methods  As mentioned in Chapter 1, the differential X-ray attenuation of various high-𝑍 elements and iodine has been extensively investigated. The following general suggestions have been made for designing studies aiming to compare the X-ray attenuation of radiopaque elements: 1) the concentrations of the radiopaque elements should be equalized, 2) the tested compounds should be dissolved in the same solvent, and 3) a wide range of X-ray tube potentials should be 18  evaluated [108]. Herein, we scanned a series of rhenium- and iodine-based solutions with an equimolar concentration of rhenium and iodine from 50 to 220 kVp.  Most equipment can only operate on a limited range of X-ray tube potentials. To scan the samples from 50 to 220 kVp, we thus had to use different instruments (and techniques). The samples were scanned from 50 to 120 kVp using a preclinical 𝜇CT scanner (eXplore CT120, TriFoil Imaging; Chatsworth, USA). They were then imaged at much higher X-ray tube potentials, from 120 and 220 kVp, using a Small Animal Radiation Research Platform (SAARP; Xstrahl; Walsall Wood, UK). This instrument can operate in two modes: CBCT and planar X-ray imaging (both of them used in our study). These two instruments are utilized for research and preclinical applications only. We were also interested in using clinical equipment, so the samples were irradiated again from 50 to 120 kVp in a clinical X-ray camera (Multix X-Ray, Siemens AG; Munich, Germany) for digital radiography.   In summary, the samples were imaged by 𝜇CT, CBCT, planar X-ray imaging, and digital radiography. The image quality was assessed by calculating the 𝐶𝑁𝑅s of the rhenium- and iodine-based solutions using data acquired with all these techniques. Using Monte Carlo, the absorbed dose was estimated for a few clinically relevant exposures.     2.2.1 Preparation of Rhenium- and Iodine-Based Solutions  A series of rhenium- and iodine-based samples were prepared using commercially available ammonium perrhenate (Sigma Aldrich) and iohexol (OmnipaqueTM, 300 mg of I per mL; GE Healthcare), which is a clinically used iodine-based XCA (see Figure 1.1). The mole percent rhenium content in ammonium perrhenate is 69.4% and the mole percent iodine content in iohexol is 46.3%. Using serial dilutions, solutions with concentrations of rhenium and iodine of 50, 100, and 200 mM were prepared.   2.2.2 Imaging of Rhenium- and Iodine-Based Solutions  2.2.2.1 Micro-Computed Tomography  The rhenium- and iodine-based solutions were pipetted into microcentrifuge tubes. The volume of each solution was 1 mL. Two additional microcentrifuge tubes with water and air were prepared. The microcentrifuge tubes with the samples were inserted in an enhanced resolution phantom (mCTP 610, Shelley Medical Imaging Technologies; London, Canada), 19  where they were arranged in a concentric circle (Figure 2.6). This phantom is made of three poly(methyl methacrylate) (PMMA, or Plexiglas®) plates contained in a cylindrical polycarbonate housing. The thickness of each plate is 1.3 cm, thereby providing a total filtration of 3.9 cm of PMMA. The phantom was utilized just to support the microcentrifuge tubes during the scans [109].     Figure 2.6. Enhanced Resolution Phantom. The phantom is constituted of three PMMA plates contained in a cylindrical polycarbonate housing with a diameter of 4.4 cm. The phantom provides a total filtration of 3.9 cm of PMMA. Model and Manufacturer: mCTP 610, Shelley Medical Imaging Technologies (London, Canada).   The samples were imaged using a preclinical 𝜇CT scanner (eXplore CT120, TriFoil Imaging; Chatsworth, USA) (Figure 2.7). According to the manufacturer’s specifications, the scanner has an inherent filtration of 1 mm of aluminum and 0.15 mm of beryllium, which corresponds to the X-ray window. As additional filtration, a 0.6 mm thick copper sheet was placed between the phantom and the X-ray tube. The samples were scanned at 50 kVp (80 mAs and 125 ms), 80 kVp (60 mA and 125 ms), and 120 kVp (40 mA and 63 ms). To evaluate the effect of increasing the amount of additional filtration on image quality, the 120 kVp scan was repeated placing 1.2 and 1.6 mm thick copper sheets between the phantom and the X-ray tube. Table 2.1 summarizes the settings of all the scans.      20    Figure 2.7. Preclinical Micro-Computed Tomography Scanner. The scanner can image samples and small animals from 50 to 120 kVp with a resolution between 25 and 100 𝜇m. The instrument is fully shielded. Model and Manufacturer: eXplore CT120, TriFoil Imaging (Chatsworth, USA). Location: Centre for High-Throughput Phenogenomics (CHTP) at the University of British Columbia (UBC)  in Vancouver, Canada.   Table 2.1. Settings for Micro-Computed Tomography. The samples were scanned at 50, 80, and 120 kVp using a preclinical 𝜇CT scanner (eXplore CT120, TriFoil Imaging; Chatsworth, USA). The current and the exposure time were chosen to avoid saturation of the detector. The spatial resolution of the images after reconstruction was 50 𝜇m.   Scan  ID X-Ray  Tube Potential (kVp) Amount of Additional Filtration (mm of Cu) Current (mA) Exposure Time (ms) Current ∙ Exposure Time Product (mAs) X01 50 0.6 80 125 10 X02 80 0.6 60 125 7.5 X03 120 0.6 40 63 2.5 X03 120 1.2 40 63 2.5 X05 120 1.6 40 125 5   The scans were performed with a rectangular field of view (FOV) with length and width of 8.5 and 5.5 cm, respectively. The acquisition consisted of 1,440 projections per scan. The scanner was operated in a step-and-shoot mode, which means that the gantry takes two projections, rotates, pauses for the duration of a pre-defined step delay, takes the next two projections, rotates, and so on. The step delay is added to cool the anode during acquisition because excessive heating in the anode causes the scan to stop before completion. A step delay of 3 s was sufficient for the 50 and 80 kVp scans. The 120 kVp scans generate more heat and thus a step delay of 7 s was required. MicroView (Parallax Innovations; Ilderton, Canada), an image processing software installed in the scanner’s computer, was used to reconstruct the data as 3D images with a matrix 21  size of 1455 x 1455 and a spatial resolution and a slice thickness of 50 𝜇m. For each 3D image, this software was also utilized to measure the mean voxel intensity of each sample in a cylindrical volume-of-interest (VOI) with diameter and height of 50 pixels. The 𝐶𝑁𝑅s of the rhenium- and iodine-based solutions were calculated relative to water using Equation 2.2. The analysis was conducted in triplicate.   2.2.2.2 Cone-Beam Computed Tomography  For this study, only the rhenium- and iodine-based solutions with a concentration of 200 mM were imaged. The solutions were scanned in plastic microcuvettes. The volume of each solution was 1 mL. Additionally, two microcuvettes with water and air were included in all the scans.  The samples were imaged using the SAARP (Xstrahl; Walsall Wood, UK) in CBCT mode. The SARRP is a research and preclinical platform which can operate between 50 and 225 kVp and is thus useful for both imaging (<140 kVp) and therapy (>140 kVp). For instance, the utilization of the SARRP for image-guided radiotherapy has been reported in mice [110,111]. Herein, we used the SARRP to scan the samples between 120 and 220 kVp. The instrument has a platform where the samples were placed in a straight line. A picture of the main components of the SARRP is shown in Figure 2.8. The only source of inherent filtration in the instrument is the X-ray window, made of 0.15 mm of beryllium. As additional filtration, a 0.15 mm thick copper sheet was placed between the samples and the X-ray tube. The samples were scanned at 120 kVp (150 𝜇A), 160 kVp (50 𝜇A), and 220 kVp (50 𝜇A). The 220 kVp scan was repeated using a 0.45 mm thick copper sheet as additional filtration. The acquisition consisted of 360 projections per scan. The gantry was moved around the sample platform at an angular speed of 6 deg s-1. A summary of the settings of all the scans is shown in Table 2.2.   22    Figure 2.8. Small Animal Radiation Research Platform. The picture shows the SARRP’s (a) detector, (b) sample platform, and (c) X-ray tube, which can operate at up to 220 kVp. Although it was developed in John Hopkins University, the SARRP is currently commercialized by Xstrahl (Walsall Wood, UK) [112]. Location: Department of Medical Physics at the BC Cancer Agency in Vancouver, Canada.   Table 2.2. Settings for Cone Beam Computed Tomography. The samples were scanned at 120, 160, and 220 kVp using the SAARP (Xstrahl; Walsall Wood, UK). The current was chosen to avoid saturation of the detector. The spatial resolution of the images after reconstruction was 320 𝜇m.    Scan  ID X-Ray  Tube Potential (kVp) Amount of  Additional Filtration (mm of Cu) Current (𝝁A) X06 120 0.15 150 X07 160 0.15 50 X08 220 0.15 50 X09 220 0.45 50   Muriplan, an image processing software developed by Xstrahl (Wallsall Wood, UK) and installed in the SARRP’s computer, was used to reconstruct the data as 3D images with a matrix size of 1024 x 1024 and a spatial resolution and a slice thickness of 320 𝜇m. A quantitative VOI-based image analysis was performed to calculate the 𝐶𝑁𝑅s of the rhenium- and iodine-based solutions. For this, ImageJ, an open-source program developed by the National Institutes of Health (NIH; Bethesda, USA), was used to calculate the mean voxel intensities of the samples in cylindrical VOIs with diameter and height of 50 pixels. The analysis was performed in triplicate.   2.2.2.3 Planar X-Ray Imaging  The samples were imaged again between 120 and 220 kVp using the SARRP, but the instrument’s mode was changed to planar X-ray imaging. Same than in the scans conducted by 23  CBCT, only the rhenium- and iodine-based solutions with a concentration of 200 mM were imaged. The solutions were irradiated in microcuvettes. The volume of each solution was 1 mL. Two microcuvettes with water and air were included in all the exposures. The samples were arranged in a straight line 30 cm away from the X-ray tube. A 0.1 mm thick copper sheet was used as additional filtration. Table 2.3 shows the settings of all the exposures.   Table 2.3. Settings for Planar X-Ray Imaging. The samples were imaged between 120 and 220 kVp using the SAARP (Xstrahl; Walsall Wood, UK). All the exposures were successfully performed with a current of 0.5 mA.   Scan ID X-Ray  Tube Potential (kVp) Amount of Additional Filtration (mm of Cu) Current (mA) X10 120 0.1 0.5 X11 140 0.1 0.5 X12 160 0.1 0.5 X13 180 0.1 0.5 X14 200 0.1 0.5 X15 220 0.1 0.5   An issue in planar X-ray imaging and digital radiography which might cause discrepancies in the pixel intensities between the left and the right sides of the images is the heel effect. This phenomenon is defined as the reduction in the intensity of the X-ray beam along the cathode-anode axis in the X-ray tube. The reduction in the X-ray beam’s intensity is caused by the steep angle of the anode, which is usually less than 17° in most equipment [113] (Figure 2.9). The photons are exposed to a differential attenuation within the target because they are produced at various depths. Generally, photons coming from greater depths within the target (i.e., the “anode side”) suffer a higher attenuation than those coming from the surface of the target (i.e., the “cathode side”) [113-115]. In other words, the probability of attenuation depends on the distance the photons travel within the target, which in turn depends on the direction at which they are emitted. To account for the heel effect, the position of the samples was changed as per indicated in Figure 2.10 and all the scans were repeated.   24    Figure 2.9. Heel Effect. As a consequence of the steep angle of the anode (𝜃), photons emitted towards the “cathode side” (designated as “𝐵” in the cartoon) have a higher energy than those emitted towards the “anode side” or perpendicular to the cathode-anode axis (designated as “𝐴” in the cartoon). That is, ℎ𝜐𝐵 > ℎ𝜐𝐴.     Figure 2.10. Setup for Planar X-Ray Imaging. The samples were imaged as per indicated in (A) (“original” configuration). Then, they were arranged as per indicated in (B) and imaged again (“flipped” configuration). The images acquired in both configurations at each X-ray tube potential were averaged to generate a single image, which was then utilized to calculate the 𝐶𝑁𝑅s of the solutions. The image shows: (a) water, (b) rhenium-based sample (200 mM), (c) iodine-based sample (200 mM), and (d) air.   Using ImageJ (NIH; Bethesda, USA), images acquired at the same X-ray tube potential but using different sample configurations (i.e., “original”, as in Figure 2.10A, and “flipped”, as in Figure 2.10B) were averaged pixel-by-pixel. All the images had a matrix with a size of 1024 x 1024. This strategy has been reported to reduce the variation in the pixel intensities on different sides of the images as a result of the heel effect [116]. To further compensate for the non-uniform intensity of the X-ray beam, the samples were kept as close to each other as possible during the scans.  The same software was also utilized to measure the mean pixel intensity of each sample in a square-shaped region-of-interest (ROI) with sides made of 50 pixels. These values were then used to calculate the 𝐶𝑁𝑅s of the rhenium- and iodine-based solutions. The analysis was performed in triplicate.    25  2.2.2.4 Digital Radiography   The rhenium- and iodine-based solutions with a concentration of 200 mM were imaged utilizing a clinical X-ray camera (Multix X-Ray, Siemens AG; Munich, Germany) (Figure 2.11). A sample holder with six 200 𝜇L wells was fabricated with PMMA to image the solutions (Figure 2.12). For all the scans, a total of 175 𝜇L of each solution were pipetted into the wells. As in all previous experiments, water and air were used as controls.     Figure 2.11. Digital Radiography Camera. The picture shows the instrument’s (a) X-ray tube and (b) detector. The equipment can operate at up to 133 kVp. Model and Manufacturer: Multix X-Ray, Siemens AG (Munich, Germany) AG. Location: Department of X-Ray/Radiology at the Leslie Diamond Health Care Centre, an institute within the Vancouver General Hospital (VGH) in Vancouver, Canada.    Figure 2.12. Sample Holder. The sample holder, made of PMMA, can accommodate up to six samples with a volume of 175 𝜇L. To make them more noticeable, the circumferences of the wells were delineated with dashed lines. There is a distance of 0.5 cm between wells. The position of the ionization chamber (10X6-0.6CT, Radcal Corporation; Monrovia, USA), as per indicated in the image, was maintained constant in all the scans.  26  The samples were irradiated at 50, 81, and 121 kVp. Due to equipment limitations, the exposures were conducted at 81 and 121 kVp instead of 80 and 120 kVp. The distance between the samples and the X-ray tube was set to 102 cm. According to the manufacturer’s specifications, the instrument provides an inherent filtration of 3.9 mm of aluminum. The exposures were carried out with varying thicknesses of additional filtration, from 0 to 1.5 mm of copper. The settings of the exposures are summarized in Table 2.4. Moreover, in all the scans, a calibrated 0.6 cm3 thimble ionization chamber (10X6-0.6CT, Radcal Corporation; Monrovia, USA) connected to a Radcal electrometer (AccuDose, Radcal Corporation; Monrovia, USA) was placed on top of the samples as per indicated in Figure 2.12 in order to measure the air-kerma.   Table 2.4. Settings for Digital Radiography. The samples were imaged between 50 and 121 kVp using a clinical X-ray camera (Multix X-Ray, Siemens AG; Munich, Germany). The current was chosen to avoid saturation of the detector.   Scan  ID X-Ray  Tube Potential (kVp) Amount of Additional Filtration (mm of Cu) Current (mA) X16 50 0 1.25 X17 50 0.1 2.5 X18 50 0.3 10 X19 50 0.5 36 X20 50 1 71 X21 50 1.5 71 X22 81 0 1.25 X23 81 0.1 1.6 X24 81 0.3 1.6 X25 81 0.5 2.2 X26 81 1 4 X27 81 1.5 4 X28 121 0 1.25 X29 121 0.1 1.25 X30 121 0.3 1.25 X31 121 0.5 1.25 X32 121 1 1.25 X33 121 1.5 1.25   All the images had a matrix with a size of 2520 x 3032. ImageJ (NIH; Bethesda, USA) was used to measure the mean pixel intensities of the samples in circular ROIs with a diameter of 5 pixels. These values were then used to calculate the 𝐶𝑁𝑅s of the rhenium- and iodine-based solutions. The analysis was performed in triplicate.  Utilizing Monte Carlo’s GATE code, which stands for Geant4/Application for Emission Tomography, the depth relative absorbed dose profile and the relative average absorbed dose in a 27  Solid Water® phantom were determined at various experimental conditions. For simplicity, the phantom was simulated as a cuboid with length, width, and height of 30, 30, and 20 cm, respectively. The calculations were performed considering an inherent filtration of 1 mm of aluminum and 0.15 mm of beryllium. The distance between the phantom and the X-ray tube was defined as 100 cm.  The simulations were conducted at 50, 80, and 120 kVp in the absence of sources of additional filtration. Then, they were repeated at 120 kVp in the presence of 0.1, 0.3, 0.5, 1, and 1.5 mm thick copper sheets. The X-ray spectra were modeled using experimental data acquired with the preclinical 𝜇CT scanner (eXplore CT120, TriFoil Imaging; Chatsworth, USA) described in Section 2.2.2.1, which was available at the moment. For each exposure, the current was adjusted in order to simulate 8.5x10-5 photons in the transmitted X-ray beam. This value was defined based on the number of photons in the transmitted X-ray beam at 120 kVp with 0.5 mm of copper as additional filtration, also determined with Monte Carlo calculations. The detector was considered to have an efficiency of 100%. All the simulations were carried out assuming a static and divergent X-ray beam with a FOV of 20 cm in diameter on the phantom’s surface. Figure 2.13 depicts a schematic of the geometry of the simulation.     Figure 2.13. Geometry of Monte Carlo Simulation. The schematic shows the X-ray beam (red), the Solid Water® phantom (blue), and the detector (brown). A FOV of 20 cm at the surface of the phantom was chosen. All the simulations were conducted assuming a total of 8.5x10-5 photons in the transmitted X-ray beam.   2.3 Results and Discussion  Our study focuses on establishing a comparison between the X-ray attenuation of rhenium and iodine from two intertwined standpoints: image quality and absorbed dose. We scanned solutions of ammonium perrhenate and iohexol with an equimolar concentration of rhenium and iodine from 50 to 220 kVp. Using preclinical equipment, the samples were imaged by 𝜇CT, 28  CBCT, and planar X-ray imaging. Using clinical equipment, they were imaged by digital radiography equipment. The 𝐶𝑁𝑅 was calculated for each solution across all scans. The absorbed dose was predicted using Monte Carlo for the exposures done by digital radiography.  The results of this work are discussed in the next few sections.   2.3.1 Imaging of Rhenium- and Iodine-Based Solutions  2.3.1.1 Micro-Computed Tomography  A series of rhenium- and iodine-based solutions with a concentration of the radiopaque element (i.e., rhenium and iodine) of 50, 100, and 200 mM were imaged by 𝜇CT at 50, 80, and 120 kVp (i.e., low-, intermediate-, and high-kVp scans, respectively).  A positive correlation was found between the 𝐶𝑁𝑅 and the concentration of the radiopaque element at each X-ray tube potential. These findings, which are consistent with other studies [117-119], are depicted graphically in Figure 2.14A-C for the scans performed with a 0.6 mm thick copper sheet. A linear regression analysis was performed to assess the relationship between these two variables (Table 2.5). A representative axial image of the samples acquired at 50 kVp is depicted in Figure 2.14D. This image clearly shows that, as the concentrations of rhenium and iodine are increased from 50 to 200 mM, there is a significant improvement in contrast. Based on the graphs shown in Figure 2.14A-C, there is a 4 to 5-fold increase in 𝐶𝑁𝑅 when the concentration is increased from 50 to 200 mM. Thus, the following discussion centers on the results when the concentrations of rhenium and iodine are 200 mM.   29    Figure 2.14. Evaluation of the Contrast-to-Noise Ratio in Micro-Computed Tomography. (A-C) The graphs show the relationship between the 𝐶𝑁𝑅 and the concentration of either rhenium or iodine in each solution. The scans were carried out at (A) 50 kVp (10 mAs), (B) 80 kVp (7.5 mAs), and (C) 120 kVp (2.5 mAs) with a 0.6 mm thick copper sheet. That is, scans X01, X02, and X03, respectively (see Table 2.1). The dashed lines represent the best linear fit to the data (see Table 2.5). (D) Axial image acquired at 50 kVp showing the following samples: iodine-based solution ([a] 200 mM, [b] 100 mM, and [c] 50 mM), rhenium-based solution ([d] 200 mM, [e] 100 mM, and [f] 50 mM), and controls ([g] water and [h] air). Attenuation: (a) 1,746 HU, (b) 863 HU, (c) 403 HU, (d) 1,297 HU, (e) 622 HU, (f) 285 HU, (g) -9.5, and (h) -994 HU. The scan corresponds to X01 (see Table 2.1). The samples were imaged by 𝜇CT (eXplore CT120, TriFoil Imaging; Chatsworth, USA).  Table 2.5. Linear Regression Analysis. The linear regression analysis was performed for the 50, 80, and 120 kVp scans conducted with a 0.6 mm thick copper sheet. That is, scans X01, X02, and X03, respectively (see Table 2.1). The linear equation has the following form: 𝐶𝑁𝑅 = m C + b, where C is the concentration of the radiopaque element, reported in mM, and m and b are the slope and the intercept of the linear equation, respectively. The coefficient of determination (𝑅2) was ~1 in all cases.   X-Ray  Contrast Agent Scan  ID X-Ray  Tube Potential (kVp) m b Ammonium Perrhenate Solution X01 50 1.3x10-2 8.2x10-2 X02 80 4.5x10-2 2.6x10-1 X04 120 5.1x10-2 2.0x10-1 Iohexol  Solution X01 50 1.7x10-2 5.4x10-2 X02 80 5.4x10-2 3.1x10-1 X04 120 4.5x10-2 3.2x10-1 30  The quantitative VOI-based analysis showed that the 𝐶𝑁𝑅s of rhenium and iodine are dependent upon the X-ray tube potential. This is exemplified in Figure 2.15 for the scans conducted with a 0.6 mm thick copper sheet. The iodine-based solution has a greater 𝐶𝑁𝑅 than the rhenium-based solution at 50 and 80 kVp. The rhenium-based solution, however, has a higher 𝐶𝑁𝑅 at 120 kVp   Figure 2.15. Effect of the X-Ray Tube Potential on the Contrast-to-Noise Ratio. The iodine-based solution exhibits a higher 𝐶𝑁𝑅 than the rhenium-based solution at 50 and 80 kVp. Contrariwise, the rhenium-based solution displays a greater 𝐶𝑁𝑅 at 120 kVp. The concentration of the radiopaque element in the solutions was 200 mM. A 0.6 mm thick copper sheet was employed as additional filtration. The samples were imaged by 𝜇CT (eXplore CT120, TriFoil Imaging; Chatsworth, USA). Settings: 50 kVp (10 mAs), 80 kVp (5 mAs), and 120 kVp (2.5 mAs). That is, scans X01, X02, and X03, respectively (see Table 2.1).   Figure 2.16 shows the effect of the addition of further sheets of copper on the 𝐶𝑁𝑅s of rhenium and iodine at 120 kVp. The difference in 𝐶𝑁𝑅 between rhenium and iodine becomes larger as the thickness of the copper sheet is increased from 0.6 to 1.6 mm. For instance, the percent 𝐶𝑁𝑅 improvement of rhenium over iodine is 14.2% when the thickness of the copper sheet is 0.6 mm, but it is 59.1% when the thickness of the copper sheet is 1.6 mm. Figure 2.17 depicts axial images acquired in all of these scans. The differences in contrast between the samples may seem subtle, but humans can only perceive a few different shades of grey. Researchers have stated that humans are able to distinguish up to 900 shades of grey, but depending on lighting conditions and people’s visual health, the range can be as low as 30 to 450 in most cases [120]. Nonetheless, images are made of pixels with a much higher number of shades of grey. The grey scale depends on the bit depth of the image: 8-bit images are made of pixels with up to 256 shades of grey (28 = 256), but 16-bit images, which are commonly used in 31  medical imaging [121-123], are made of pixels with up to 65,536 shades of grey (216 = 65,536). Furthermore, high levels of noise considerably impact the ability to discern one shade of grey from another [124]. The calculation of 𝐶𝑁𝑅s is thus a useful strategy to highlight differences in contrast between materials of interest, providing even a better understanding of the parameters which require optimization in order to increase image quality.    Figure 2.16 Effect of the Amount of Additional Filtration on the Contrast-to-Noise Ratio. The percent 𝐶𝑁𝑅 improvement of rhenium over iodine is enhanced as the amount of addition filtration is increased. The concentration of the radiopaque element in the solutions was 200 mM. The samples were imaged by 𝜇CT (eXplore CT120, TriFoil Imaging; Chatsworth, USA). All the scans were conducted at 120 kVp. They correspond to scans X03, X04, and X05 (see Table 2.1).     32    Figure 2.17. Micro-Computed Tomography Imaging of Rhenium and Iodine. Representative axial images acquired at 120 kVp by 𝜇CT (eXplore CT120, TriFoil Imaging; Chatsworth, USA) with copper sheets with the following thicknesses: (A) 0.6, (B) 1.2, and (C) 1.6 mm. The scans correspond to X03, X04, and X05, respectively (see Table 2.1)  The images depict the following samples: rhenium-based solutions ([a] 200 mM, [b] 100 mM, and [c] 50 mM), iodine-based solutions ([d] 200 mM, [e] 100 mM, and [f] 50 mM), and controls ([g] water and [h] air). Attenuation: In (A): (a) 917 HU, (b) 460 HU, (c) 210 HU, (d) 803 HU, (e) 393 HU, (f) 176 HU, (g) 0, and (h) -998 HU. In (B): (a) 942 HU, (b) 473 HU, (c) 218 HU, (d) 661 HU, (e) 321 HU, (f) 138 HU, (g) 1, and (h) -996 HU. In (C): (a) 971 HU, (b) 486 HU, (c) 227 HU, (d) 610 HU, (e) 305 HU, (f) 137 HU, (g) 0, and (h) -998 HU.  The results discussed above indicate that rhenium exhibits a kVp-dependent superiority in 𝐶𝑁𝑅 (and thus, in X-ray attenuation) over iodine. The higher X-ray attenuation of rhenium at 120 kVp is related to the differential K shell binding energy, or K-edge, between rhenium and iodine. The K-edge is defined as a sharp rise in the linear X-ray attenuation coefficient (𝜇𝑋) which occurs at a discreet energy just above the binding energy of the electrons in the K shell of an atom [125]. To put it differently, it represents the energy required to eject an electron from an atom’s K shell. Similar increases in the linear X-ray attenuation coefficient are observed for all the other electron shells (L, M, N, and so on), but they take place at much lower energies, edges [126,127].  The increase in the linear X-ray attenuation coefficient at each of the absorption edges is caused by the photoelectric effect (see Section 2.1.2). The probability of occurrence of an interaction between an incident photon and an inner shell electron through the photoelectric effect is described by the following expression:   𝜏 ∝ 𝑍3/(ℎ𝜐)3          (Equation 2.4)  where 𝜏 is the contribution of the photoelectric effect to the linear X-ray attenuation coefficient, specifically known as the photoelectric effect coefficient [57]. For example, when an incident 33  photon has energy just above the binding energy of an electron in the L shell, the probability of occurrence of a photoelectric interaction involving the L shell becomes very high (i.e., the process is energetically more favorable). According to Equation 2.4, beyond this point, the probability decreases approximately at a rate of 1/(ℎ𝜐)3 until the next “jump” or “discontinuity”, the K-edge. Moreover, Equation 2.4 shows that the photoelectric effect coefficient depends on the atomic number raised to the third power. For this reason, high-𝑍 elements are of special interest in the search of novel XCAs.  The National Institute of Standards and Technology (NIST) has published a series of tables with the mass X-ray attenuation coefficient (i.e., the linear X-ray attenuation coefficient divided by the density, expressed in units of cm2 g-1) as a function of the photon energy. These tables are available for elements with an atomic number between 1 and 92 and 48 compounds and materials of interest (e.g., water, air, bone, soft tissue, etc.) and they include all the absorption edges between 1 keV and 20 MeV [128]. Figure 2.18 graphically illustrates the change in the mass X-ray attenuation coefficient of rhenium and iodine at clinically relevant photon energies. The graph shows that the K-edge of rhenium occurs at 71.7 keV, whereas the K-edge of iodine occurs at 33.1 keV. Above the K-edge of rhenium, the mass X-ray attenuation coefficient of rhenium is consistently higher than iodine. This in turn suggests that rhenium should display an enhanced X-ray attenuation above 71.7 keV.    Figure 2.18. Effect of the Photon Energy on the Mass X-Ray Attenuation Coefficient. The graph depicts the variation in the mass X-ray attenuation coefficient of rhenium, iodine, water, and air between 3 and 150 keV. The K-edges of rhenium and iodine are shown, which occur at 71.7 and 33.1 keV, respectively. Based on data available from the NIST [128].   34  SpekCalc is a computational tool available in MatlabTM to predict X-ray spectra between 30 and 140 kVp [129,130]. SpekCalc requires information about: 1) the characteristics of the X-ray tube (i.e., target material, anode angle, sources of inherent filtration) and 2) the settings of the scan (e.g., X-ray tube potential, current, exposure, sources of additional filtration). This software was utilized to predict the X-ray spectra for the scans performed at 120 kVp, which is the maximum X-ray tube potential that can be supplied to the X-ray tube of the eXplore CT120 scanner (TriFoil Imaging; Chatsworth, USA). Figure 2.19 depicts the 120 kVp X-ray spectra, where it can be seen that the X-ray beam produced in each scan was significantly hardened with the addition of copper sheets. For instance, the mean energy of the 120 kVp X-ray beam was increased from 39.8 keV (without a copper filter) to 63.2, 67.8, and 69.5 keV (with 0.6, 1.2, and 1.6 mm thick copper sheets, respectively). According to Figure 2.18, iodine exhibits an enhanced X-ray attenuation from 33.1 to 71.7 keV. In other words, iodine is more likely than rhenium to interact with low-keV photons. Thus, the main factor contributing to the increased 𝐶𝑁𝑅 of rhenium at 120 kVp is the removal of a great proportion of the low-keV photons by the copper sheets.    Figure 2.19. Effect of Copper on the 120 kVp X-Ray Spectra. The image shows the 120 kVp X-ray spectra transmitted through copper filters with thicknesses up to 1.6 mm. The X-ray spectra were predicted using SpekCalc.   Furthermore, Figure 2.19 shows that the mean energy of the 120 kVp X-ray beam is below the K-edge of rhenium even after the addition of a 1.6 mm copper sheet. To determine if it was possible to further increase the difference in 𝐶𝑁𝑅 between rhenium and iodine, the rhenium- and iodine-based solutions were imaged at much higher X-ray tube potentials, from 120 to 220 kVp. This was possible using the SARRP (Xstrahl; Walsall Wood, UK), which can operate in two 35  modes: CBCT and planar X-ray imaging. As a rule of thumb, the mean energy of the X-ray beam is approximately a third of the maximum energy (which is equivalent to the X-ray potential) [57]. Hence, without taking into account sources of additional filtration, a 220 kVp X-ray beam has a mean energy of ~73 keV, which is slightly above the K-edge of rhenium. The results of these experiments are discussed in the next two sections.   2.3.1.2 Cone Beam Computed Tomography  Figure 2.20 depicts the variation in the 𝐶𝑁𝑅s of the rhenium and iodine-based solutions at 120, 160, and 220 kVp. The scans were conducted with a 0.15 mm thick copper sheet as additional filtration. Due to the increase in the number of incident photons, both solutions display greater 𝐶𝑁𝑅s as the X-ray tube potential is increased. As expected, rhenium has a higher 𝐶𝑁𝑅 than iodine. Additionally, when the thickness of the copper sheet is increased to 0.45 mm at 220 kVp, the difference in 𝐶𝑁𝑅 between rhenium and iodine is increased significantly (Figure 2.21). The percent 𝐶𝑁𝑅 improvement of rhenium over iodine is 13.2% with the use of a 0.15 mm thick copper sheet, but it is 36.9% with the use of a 0.45 mm thick copper sheet.    Figure 2.20. Effect of the X-Ray Tube Potential on the Contrast-to-Noise Ratio. The rhenium-based solution has a higher 𝐶𝑁𝑅 than the iodine-based solution from 120 to 220 kVp. The concentration of the radiopaque element in the solutions was 200 mM. A 0.15 mm thick copper sheet was employed as additional filtration. The samples were imaged by CBCT using the SARRP (Xstrahl; Walsall Wood, UK). Settings: 120 kVp (150 𝜇A), 160 kVp (50 𝜇A), and 220 kVp (50 𝜇A). That is, scans X06, X07, and X08, respectively (see Table 2.2).  36   Figure 2.21. Effect of the Amount of Additional Filtration on the Contrast-to-Noise Ratio. The percent 𝐶𝑁𝑅 improvement of rhenium over iodine is enhanced as the amount of additional filtration is increased. The concentration of the radiopaque element in the solutions was 200 mM. The samples were imaged by CBCT using the SARRP (Xstrahl; Walsall Wood, UK). The scans were performed at 220 kVp with a current of 50 𝜇A. That is, scans X08 and X09 (see Table 2.2).   These results are in agreement with what was observed when the samples were imaged by 𝜇CT: 1) the 𝐶𝑁𝑅 of rhenium is higher than iodine at 120 kVp and 2) the percent 𝐶𝑁𝑅 improvement of rhenium over iodine becomes larger when copper sheets with increasing thicknesses are utilized. As demonstrated by imaging the samples by CBCT, these findings are consistent at much higher X-ray tube potentials, up to 220 kVp.  A fundamental consideration in X-ray imaging revolves around acquiring useful images for assessing the patient’s clinical picture and making a diagnosis while restricting at the same time the absorbed dose as much as possible. Generally, low-keV photons are absorbed by the body more easily than high-keV photons and have a minimal contribution to the generation of an image [131]. After passing through the patient, the X-ray beam carries a pattern of intensity that is dependent upon the composition, size, and location of different tissues and organs in the body [132-134]. Although the selection of the X-ray tube potential is governed by the sensitivity of specific body structures to radiation, it is based on experience for the most part. For example, imaging of thinner regions of the body (e.g., arms, hands, and feet) is done between 50 and 60 kVp, whereas imaging of thicker, more attenuating regions of the body (e.g., lumbar spine, pelvis, and skull) is done between 70 and 90 kVp (typically by digital radiography) [95]. However, high-kVp scans, between 80 and 140 kVp, are routinely used in whole-body and chest CT scans of average-size and large patients [135,136].  37  Our study has provided evidence of the usefulness of rhenium in X-ray imaging. Furthermore, we have demonstrated that the 𝐶𝑁𝑅 of rhenium is significantly higher than iodine in high-kVp scans even in the presence of large thicknesses of copper sheets. These findings suggest that rhenium could be used as an XCA, particularly in high-kVp scans. This has important implications from a clinical standpoint. A significant amount of copper could be added to remove the majority of low-keV photons, which are mostly unproductive to generate an image but contribute to increase the absorbed dose unnecessarily [131]. This approach has the potential of reducing absorbed dose while maintaining image quality, which is the main trade-off in X-ray imaging [95,96].     2.3.1.3 Planar X-Ray Imaging  The samples were imaged again between 120 and 220 kVp with the SARRP, but in X-ray imaging mode. The relationship between the 𝐶𝑁𝑅 of the rhenium- and iodine-based solutions and the X-ray tube potential is depicted in Figure 2.22. The graph shows that the 𝐶𝑁𝑅 of iodine at 220 kVp is 26.4% greater than at 120 kVp. This moderate improvement in 𝐶𝑁𝑅 is attributable to the overall increase in photon count at 220 kVp. Rhenium, on the other hand, shows an increase in 𝐶𝑁𝑅 of 39.0%. As mentioned at the end of Section 2.3.1.2, the peak maximum in a typical 220 kVp X-ray spectrum occurs at ~73 keV. The significant improvement in 𝐶𝑁𝑅 of rhenium at 220 kVp is thus associated with the increase in the proportion of photons with energy above the K-edge of rhenium. Furthermore, like in our two previous experiments, it was found that the difference in 𝐶𝑁𝑅 between rhenium and iodine is enhanced when the X-ray tube potential is increased. For example, the percent 𝐶𝑁𝑅 improvement of rhenium over iodine is 32.2% at 120 kVp, but it is 45.3% at 220 kVp.   38   Figure 2.22. Effect of the X-Ray Tube Potential on the Contrast-to-Noise Ratio. The rhenium-based solution consistently displays a greater 𝐶𝑁𝑅 than the iodine-based solution from 120 to 220 kVp. The percent 𝐶𝑁𝑅 improvement of rhenium over iodine is enhanced as the X-ray tube potential is increased. The concentration of the radiopaque element in the solutions was 200 mM. The samples were imaged by planar X-ray imaging using the SARRP (Xstrahl; Walsall Wood, UK). All the exposures were carried out with a current of 50 𝜇A and 0.1 mm of copper as additional filtration. That is, scans X10 to X15 (see Table 2.3).      The majority of currently utilized equipment for X-ray imaging, including modern dual energy CT scanners and dual energy digital radiography cameras, cannot operate above 140 kVp [137-142]. Herein, though, exposures up to 220 kVp were performed in order to explore thoroughly the kVp-dependent superiority in X-ray attenuation of rhenium over iodine. Taking into account the aforementioned restriction in X-ray tube potential, the samples were imaged at 50, 81, and 121 kVp using a modern, clinical X ray camera. For this experiment, both image quality and radiation dose were evaluated. As in the previous experiments, the 𝐶𝑁𝑅s of the rhenium- and iodine-based solutions was calculated. However, using an ionization chamber, the air-kerma was measured in all the exposures. Additionally, the absorbed dose was estimated with Monte Carlo calculations.     2.3.1.4 Digital Radiography  The variation in 𝐶𝑁𝑅 at 50, 81, and 121 kVp as a function of the amount of additional filtration is graphically shown in Figure 2.23. The experiments conducted by 𝜇CT showed that iodine exhibits a greater 𝐶𝑁𝑅 than rhenium at 50 and 80 kVp (see Figure 2.15). Differences in X-ray attenuation are expected when the thicknesses of the copper sheets are varied because they change the energy spectrum of the X-ray beam incident on the samples. This is illustrated in Figures 2.23A and B, where it is seen that rhenium can exhibit a greater 𝐶𝑁𝑅 than iodine in low- 39  and intermediate-kVp scans depending on the amount of additional filtration. Specifically, at 50 kVp, rhenium has a higher 𝐶𝑁𝑅 than iodine in two cases: 1) without any source of additional filtration and 2) with a 0.1 mm thick copper sheet (Figure 2.23A). Although the differences between the two solutions are smaller than at 50 kVp, at 81 kVp, rhenium displays a higher 𝐶𝑁𝑅 than iodine in three cases: 1) without any source of additional filtration, 2) with a 0.1 mm thick copper sheet, and 3) with a 1.5 mm thick copper sheet (Figure 2.23B).     Figure 2.23. Evaluation of the Contrast-to-Noise Ratio in Digital Radiography. (A) 50 kVp, (B) 80 kVp, and (C) 120 kVp. The concentration of the radiopaque element in the formulations was 200 mM. The samples were imaged by digital radiography using a clinical X-ray camera (Multix X-Ray, Siemens AG; Munich, Germany). The scans correspond to (A) X16 to X21, (B) X21 to X27, and (C) X30 to X33 (see Table 2.5). The images from scans X28 and X29 were excluded from the analysis due to the lack of contrast. As a consequence of the use of insufficient additional filtration in these two scans, the detector was saturated with photons when the current was 1.25 mA, which is the lowest current which can be selected in the instrument.   Similar to the scans performed by 𝜇CT, CBCT, and planar X-ray imaging, rhenium consistently has a greater 𝐶𝑁𝑅 than iodine at 121 kVp (Figure 2.23C). As observed before, while the 𝐶𝑁𝑅𝑠 of both solutions decrease as the thickness of the copper attenuator increases, the 40  difference in 𝐶𝑁𝑅 between rhenium and iodine actually increases. For example, the percent 𝐶𝑁𝑅 improvement of rhenium over iodine is 19.8% in the presence of a 0.3 mm thick copper sheet, but it is 75.9%, almost four times higher, in the presence of a 1.5 mm thick copper sheet.  The air-kerma measured in each exposure is reported in Table 2.6. The mean energy of the X-ray beam produced at each combination of X-ray tube potential and amount of additional filtration, calculated from X-ray spectra predicted using SpekCalc, is also shown in Table 2.6. A significant degree of beam hardening is observed at the three X-ray tube potentials, but the effect is more prominent at 50 kVp. To put it into perspective, without additional filtration, at 50 kVp, the X-ray beam’s mean energy is slightly more than half the maximum energy, or specifically, 57% of 50 keV. Nonetheless, upon addition of a 1.5 mm thick copper sheet, the mean energy is equivalent to 90% the maximum energy, which corresponds to an increase of 33%. Conversely, at 81 kVp, the increase in mean energy is 25%, whereas at 121 kVp, it is only 22%.   Table 2.6. Measurement of the Air-Kerma and the Mean Energy of the X-Ray Beam. The air-kerma is reported as the mean of three independent measurements. The standard deviation is zero in cases where it is not reported. The X-ray spectra were predicted with SpekCalc taking into consideration the settings of the exposures.   Scan  ID X-Ray Tube Potential (kVp) Amount of Additional Filtration (mm of Cu) Air-Kerma (𝝁Gy) Mean Energy (keV) X17 50 0 14.9 ± 0.1 28.5 X18 50 0.1 10.5 36.2 X19 50 0.3 11.2 39.6 X20 50 0.5 15.4 41.4 X21 50 1 4.2 43.9 X22 50 1.5 1.3 45.1 X23 81 0 49.3 ± 0.2 43.1 X24 81 0.1 31.9 ± 0.6 48.1 X25 81 0.3 15.4 ± 0.1 53.4 X26 81 0.5 12.5  56.5 X27 81 1 9.3 61.0 X28 81 1.5 21.2 ± 0.1 63.7 X29 121 0 114.9 ± 0.2 53.9 X30 121 0.1 74.0 ± 0.2 59.7 X31 121 0.3 46.0 ± 0.1 65.9 X32 121 0.5 33.4 ± 0.1 69.8 X33 121 1 19.7 ± 0.1 76.1 X34 121 1.5 11.7 ± 0.1 80.5   The dissimilar degree of beam hardening between the low-, intermediate-, and high-kVp scans is attributed to the positive correlation between the X-ray tube potential and the energy spectrum of the X-ray beam. At 50 kVp, the X-ray beam is constituted of photons with energy up 41  to 50 keV, which are very likely to be absorbed. As the number of photons absorbed in the copper sheet is increased, the X-ray beam’s energy spectrum is decreased dramatically (and thus, the transmitted X-ray beam is predominantly constituted of high-keV photons). The number of photons is markedly reduced as well. To compensate for the loss in photon count, especially at 50 kVp, the current was significantly raised as the thickness of the copper sheet was augmented. As stated in Table 2.6, at this X-ray tube potential, without additional filtration, the current was set to 1.25 mA, but with a 1.5 mm thick copper attenuator, it was set to 71 mA. Contrariwise, at 121 kVp, the current was maintained at 1.25 mA in all the irradiations.  The substantial shift in the mean energy of X-ray beam generated at 50 kVp, and even at 81 kVp, changes the penetrability of the X-ray beam. This in turn has an effect on the interaction of the photons with the samples. From Table 2.6, it is observed that the X-ray beam of a high-filtered (≥1 mm of Cu) 50 kVp exposure can have a higher mean energy than the X-ray beam of a low-filtered (<0.1 mm of Cu) 81 kVp exposure. For example, in the 50 kVp irradiation with 1.5 mm of copper as additional filtration, the X-ray beam’s mean energy was found to be 45.1 keV, but in the 80 kVp irradiation with no additional filtration, it was found to be 43.1 keV. A similar trend is seen between high-filtered (≥1 mm of Cu) 81 kVp exposures and low-filtered (≤0.1 mm of Cu) 121 kVp exposures. These findings explain why, with the utilization of a large amount of copper at 81 kVp, rhenium displays a greater 𝐶𝑁𝑅 than iodine.  The high 𝐶𝑁𝑅 of rhenium in the low-filtered 50 and 81 kVp exposures is attributable to the accentuation of the photoelectric effect at the L1-edge of rhenium, which occurs at 12.5 keV [128]. The subscript indicates the type of electron involved in the transition. As a rule, only 1s electrons can be excited at the K-edge. However, either 2s or 2p electrons can be excited at the next absorption edge, the L-edge. The excitation of a 2s electron is specifically known as the L1-edge. The excitation of a 2p electron, on the other hand, is split into two absorption edges, L2 and L3, due to the spin-orbit coupling energy of the 2p5 configuration which is created when a 2p electron is excited. Due to the difference in energy between 2s and 2p electrons, the L1-edge takes place at a higher energy [143]. Thus, rhenium’s L2- and L3-edges occur at 12.0 and 10.5 keV, respectively. Moreover, iodine’s L1-, L2-, and L3-edges occur 5.2, 4.8, and 4.6 keV, respectively [128]. Figure 2.18, first discussed in Section 2.3.1.1, also depicts the sharp-increase in the mass X-ray attenuation coefficient of rhenium between 10 and 12 keV as a result of its L1-, 42  L2-, and L3-edges. The graph shows that rhenium has a greater mass X-ray attenuation coefficient than iodine below the K-edge of iodine 33.1 keV.  To compare the differential X-ray attenuation of rhenium and iodine across all the exposures, we calculated a rhenium-to-iodine X-ray attenuation ratio. This was accomplished by dividing the mean pixel intensity of rhenium by the mean pixel intensity of iodine and dividing this quotient by the air-kerma measured in each scan. Figure 2.24 illustrates the relationship between the rhenium-to-iodine X-ray attenuation ratio, determined experimentally, and the mean energy of the X-ray beam. As a reference, a theoretical rhenium-to-iodine mass X-ray attenuation coefficient ratio was also calculated and plotted in Figure 2.24. Basically, when the rhenium-to-iodine X-ray attenuation ratio is above 1, rhenium outperforms iodine, but when it is below 1, iodine outperforms rhenium. The deviation between the experimental and the theoretical curves is expected for a number of reasons; for example, scattering from the surrounding media and levels of noise in the images.    Figure 2.24. Rhenium-to-Iodine X-Ray Attenuation Ratio. The experimental curve was derived by dividing the mean pixel intensity of rhenium by the mean pixel intensity of iodine and dividing this value by the air-kerma. The theoretical curve, on the other hand, was calculated from data reported by the NIST [128]. Basically, it represents the ratio of the mass X-ray attenuation coefficient of rhenium to the mass X-ray attenuation of iodine.   From Figure 2.24, it is evident that the X-ray attenuation of rhenium is favored when the mean energy of the X-ray beam is slightly above 71.7 keV, that is, rhenium’s K-edge. This in fact supports of all our previous experiments. The enhanced X-ray attenuation of rhenium observed on the left-section of the graph provide further evidence of the 𝐶𝑁𝑅 superiority of the rhenium-based solution in the low-filtered 50 and 81 kVp scans. Furthermore, it is interesting that, at the experimental conditions where iodine is expected to exhibit an improved X-ray 43  attenuation, a substantial difference between the two solutions was not observed. For instance, the experimental rhenium-to-iodine X-ray attenuation ratio is fairly close to 1 between 40 and 60 keV. This in turn indicates that even in these scans the X-ray attenuation of rhenium was comparable to iodine.  A higher air-kerma is expected when a larger X-ray tube potential is applied because photons with greater energy are produced. The air-kerma, however, does not necessarily correlate with absorbed dose [144]. Using Monte Carlo, we appraised the differences in absorbed dose in 50, 80, and 120 kVp exposures, similar to the ones described above. The absorbed dose was calculated for a 30 x 30 x 20 cm Solid Water® phantom (see Figure 2.13). The current was adjusted in order to simulate 8.5x10-5 photons in the transmitted X-ray beam. Considering that the detector was assumed to have an efficiency of 100%, this strategy ensures that, irrespective of the X-ray tube potential and the amount of additional filtration, the scans generate images with similar quality and comparable noise.  Figure 2.25A depicts the change in the relative average absorbed dose per slice as a function of the depth of the phantom (i.e., the depth absorbed dose profile) for 50, 80, and 120 kVp exposures performed with no additional filtration. The graph shows that, at 50 kVp, significantly more radiation dose is deposited in the phantom. As a matter of fact, the relative average absorbed dose was found to be five times larger at 50 kVp than at 120 kVp. Also, while the difference is not as large, the relative average absorbed dose at 80 kVp is almost double than at 120 kVp. These values are summarized in Table 2.7. Figure 2.25B shows that the relative absorbed dose is significantly reduced at 120 kVp when the amount of additional filtration is increased. For example, the relative average absorbed dose is reduced 37.6% after the addition of a 1.5 mm thick copper sheet. The curves show that the highest deposition of radiation dose occurs within a couple of centimeters of the surface of the phantom. This is referred to as the build-up region. This phenomenon is typically attributed to electrons set in motion by scattered photons following a Compton interaction [145,146]. These results indicate that, upon optimization (e.g., current, exposure time, and addition of copper sheets), it may be feasible to reduce the absorbed dose in high-kVp irradiations while allowing the acquisition of a useful image at the same time.   44    Figure 2.25. Depth Relative Absorbed Dose Profiles. Using Monte Carlo, the depth relative absorbed dose profiles were predicted for exposures carried out (A) at 50, 80, and 120 kVp without additional filtration and (B) at 120 kVp with copper sheets with thicknesses between 0.1 and 1.5 mm. The phantom was simulated as a 30 x 30 x 20 cm Solid Water® cuboid. The current was adjusted in order to simulate 8.5x10-5 photons in the transmitted X-ray beam (i.e., the number of photons produced at 120 kVp in the presence of a 0.5 mm thick copper sheet).  Table 2.7. Relative Average Absorbed Dose. Using Monte Carlo, the relative average absorbed dose in a 30 x 30 x 20 cm Solid Water® cuboid was calculated at 50, 80 and 120 kVp. The effect of the addition of copper sheets with thicknesses between 0.1 and 1.5 mm was also investigated. The relative average absorbed dose was considered to be 1 at 120 kVp in the presence of a 0.5 mm thick copper sheet. The table also shows the mean energy of the X-ray beam incident on the phantom, which are similar to the experimental values within 10% (see Table 2.6). The relative current is the scaling factor utilized to equalize the number of photons in the transmitted X-ray beam, which was set to 8.5x10-5 (i.e., the number of photons produced at 120 kVp in the presence of a 0.5 mm thick copper sheet).   X-Ray Tube Potential (kVp) Amount of Additional Filtration (mm of Cu) Mean Energy (keV) Relative Current Relative Average Absorbed Dose 50 0 31.8 10.2 7.0 80 0 42.4 2.8 2.2 120 0 53.7 1.6 1.4 120 0.1 59.0 1.3 1.2 120 0.3 64.9 1.1 1.1 120 0.5 68.1 1.0 1.0 120 1 74.1 0.9 0.9 120 1.5 78.4 0.8 0.9   2.4 Conclusions  This study showed that enhanced contrast can be achieved with the utilization of a rhenium-based solution, particularly in high-kVp scans. The X-ray attenuation of rhenium was compared with iodine using various techniques. The parameters that were varied the scans were the X-ray tube potential and the amount of additional filtration, defined by the thickness of a copper sheet. These two parameters determine the mean energy of the X-ray beam.  45  At an equimolar concentration of the radiopaque element, by 𝜇CT, it was found that a rhenium-based solution exhibits a greater 𝐶𝑁𝑅 than an iodine-based solution at 120 kVp. By CBCT, it was confirmed that this trend prevails at up to 220 kVp. Furthermore, within this range, the percent 𝐶𝑁𝑅 improvement of rhenium over iodine is enhanced as the amount of additional filtration is increased. For example, the difference in 𝐶𝑁𝑅 between rhenium and iodine is more than double when the thickness of the copper attenuator is increased from 0.15 to 0.45 mm at 220 kVp. The consistency of these findings was corroborated by planar X-ray imaging. Using a clinical X-ray camera, the solutions were imaged by digital radiography. This study revealed that, provided the X-ray beam is sufficiently hardened, rhenium can also exhibit a greater 𝐶𝑁𝑅 than iodine below 120 kVp.  Altogether, our study showed that rhenium displays a kVp-dependent superiority in X-ray attenuation over iodine. In high-kVp scans, the enhanced X-ray attenuation of rhenium is the result of the following two phenomena: 1) the increase likelihood of photoelectric interactions above the K-edge of rhenium (i.e., 71.3 keV) and 2) the increase likelihood of Compton interactions when the X-ray beam has a high proportion of high-keV photons.  Ideally, it is fundamental to achieve adequate image quality with a reasonably low radiation dose per image. We demonstrated that the absorbed dose in high-kVp can be significantly reduced by removing a great proportion of the low-keV photons without compromising image quality. Through Monte Carlo calculations, it was shown that, in the presence of a 1.5 mm thick copper attenuator, the absorbed dose in a 120 kVp exposure can be reduced by 85% relative to a 50 kVp exposure when images with similar quality and comparable noise are theoretically acquired. This in turn suggests that high-kVp scans might impose a radiation dose advantage because low-keV photons, which are more likely to be absorbed by the patient, can be easily removed without compromising the quality of the images.   46  Chapter 3 Development and Imaging of Radiopaque Electrospun Scaffolds  The previous chapter provided evidence of the kVp-dependent superiority in X-ray attenuation of rhenium relative to iodine. At an equimolar concentration, rhenium has a higher X-ray attenuation (and thus, it shows more contrast) than iodine. These results were consistently observed in high-kVp scans, between 120 and 220 kVp. Therefore, we demonstrated rhenium’s usefulness in X-ray imaging. This chapter describes a study conducted to exploit the X-ray attenuation of rhenium, along with its chemical versatility, in a very specific application within the field of X-ray imaging.  A rhenium-doped scaffold made of polycaprolactone (PCL) (a fully biodegradable and biocompatible polymer) was produced by electrospinning (a technique to make scaffolds constituted of nanofibers, or NFs). This radiopaque electrospun scaffold was utilized to coat catheters. Typically, catheters are placed inside the body under X-ray imaging guidance. Large doses of I-XCAs are administered to patients in catheterizations because the majority of catheters are radiolucent. The incorporation of the aforementioned scaffold onto the surface of the catheters made them radiopaque, which will provide a visual guide for physicians during catheterizations, potentially reducing or avoiding altogether the use of I-XCAs.   3.1 Background  This section is divided into two parts. The first part focuses on the theory behind electrospinning with particular emphasis on the parameters and properties that affect the structure of the scaffolds. This technique has been demonstrated to reliably produce polymer-based scaffolds made of NFs. The second part revolves around the current state of the art in the development and utilization of radiopaque catheters, a field of research which has not seen significant improvement over the past few decades.   3.1.1 Principles and Applications of the Electrospinning Technique  Also known as electrostatic spinning, electrospinning is a technique capable of generating polymer-based scaffolds made of fibers with a diameter in the nanometer range: that is, NFs [147]. The first patent describing the underlying principles of electrospinning was issued by the United States Patent and Trademark Office over 80 years ago, in 1934: “Process and Apparatus 47  for Preparing Artificial Threads” [148]. Nevertheless, it was during the decades of the 1980s and 1990s when the potential of this technology became indisputable. Around this time, a variety of scaffolds were fabricated from solutions containing a wide range of polymers: polyesters (e.g., polyethylene therephthalate, usually referred to as PET), polyolefines (e.g., polyethylene and polypropylene), and polyamides (e.g., wool, silk, and aramid), to name but a few [149,150].  Electrospinning is a straightforward technique which uses an external electric field to produce and collect NFs on a plate. Altogether, it involves two main processes: 1) the formation of the Taylor cone and 2) the bending instability of the jet. A typical electrospinning setup requires drawing up a polymer-based solution into a syringe assembled with a blunt needle which functions as a capillary. The solution is held at the end of the capillary due to surface tension forces. Upon assembling, a voltage is applied in order to generate an electric field and charge the solution. Ultimately, the charges in the solution result in repulsive electric forces directly opposite to the surface tension forces [151]. As the intensity of the electric field is increased, ions in the solution start to aggregate causing the hemispherical surface at the tip of the capillary to elongate in a conical shape, referred to as the Taylor cone [152,153].  A charged jet is ejected from the tip of the Taylor cone when the electric field reaches a critical value at which the repulsive electric forces overcome the surface tension forces [154]. As the jet approaches to the collecting plate, the surface charge density in the solution increases. Interestingly, when the surface charge density becomes sufficiently high, the jet starts to twist and eventually adopts a conical trajectory. This phenomenon is often termed bending instability and it causes the jet to stretch to the point that its diameter is at the nanoscale. During this process, the solvent evaporates, leaving behind NFs which lay themselves randomly on the collecting plate [155,156]. This series of events is depicted in Figure 3.1.   48    Figure 3.1. Schematic of the Electrospinning Technique. During electrospinning, a voltage is applied to charge a polymer-based solution contained in a capillary. A charged jet is ejected when the surface tension forces are overcome by the repulsive electric forces at the tip of the capillary. On its way to the collecting plate, which is made of a conducting material, the jet is twisted and stretched. As the solvent evaporates, NFs are produced. The image illustrates: 1) the formation of the Taylor cone and 2) the bending instability of the jet.   After deposition on the collecting plate, the NFs gradually form scaffolds that have found interesting applications in diverse disciplines due to their extremely high surface area-to-volume ratio. For instance, since electrospinning was shown to create scaffolds in a controlled and reproducible way, it rapidly became the focus of attention in regenerative medicine research. The NFs have a similar range of diameters than the macromolecules found in the extracellular matrix (ECM), like collagen [157,158]. The ECM is a complex network of fibrous proteins and proteoglycans which provides structural and biochemical support to the cells. In general, tissues are considered an agglomeration of tightly packed cells, but their volume is partly made up of extracellular space which is ultimately filled by the ECM [159]. Extensive work in this field has shown that: 1) the structure of the NF network mimics the hierarchical organized fibrous mesh characteristic of the ECM and 2) the high surface area-to-volume ratio of the NFs facilitates cell attachment and drug loading [160-162].  In general, electrospinning has found important applications in tissue engineering [163-165], but electrospun scaffolds have also been utilized in food science (e.g., filtration, high-performance packaging materials, immobilization of bioactive compounds: enzymes, vitamins, minerals, and antimicrobials) [166-168], energy and environmental science (e.g., batteries, solar cells) [169-171], textiles (e.g., resistant fabrics) [172-174], and cosmetics (e.g., deodorants, antiperspirants, skin functional textiles, chemical care products) [175,176]. Due to its simplicity 49  and versatility, which is corroborated by the feasibility of using various natural and synthetic polymers, electrospinning is presently the preferred technique to fabricate scaffolds [177,178].  PCL is one of the leading polymers in the fabrication of scaffolds as a result of its biodegradable and biocompatible nature [179,180]. This polymer is a semi-crystalline linear aliphatic polyester susceptible to undergo autocatalyzed bulk hydrolysis and slow degradation in vivo [181,182]. PCL has a low melting temperature (𝑇𝑚; 56 to 64 °C) and very low glass transition temperature (𝑇𝑔; ≈-60 °C). The production of scaffolds with PCL is relatively easy because it exhibits a large viscoelasticity at room temperature. Briefly, viscoelasticity refers to the property of materials which exhibit both viscous and elastic behaviors when stress is applied [183]. In other words, viscoelastic materials show a time-dependent (viscous component) strain (elastic component) when they are stretched. The viscoelasticity of PCL can be quantified in terms of its tensile strength and Young’s modulus, both of which have a wide range: 4 to 785 MPa and 0.21 to 0.44 GPa, respectively [184,185].  For the above-mentioned reasons, PCL has been used quite extensively to produce scaffolds by many different techniques, including: solvent casting [186,187], rapid prototyping [188,189], thermally induced polymer phase separation [190], and melt molding [191]. In this study, we are interested in making PCL-based scaffolds by electrospinning. This technique allows the production of scaffolds in a controlled and reproducible way simply by adjusting a few parameters (e.g., electrospinning distance and voltage). We evaluated the utilization of both 10 and 45 kDa PCL (i.e., low- and intermediate-molecular weight PCL, respectively). Both polymers are readily commercially available through Sigma Aldrich.   3.1.2 X-Ray Imaging Guidance in Catheterizations  Currently, vascular and dilatation (i.e., with an expandable balloon) catheters are routinely used in a variety of diagnostic and therapeutic procedures. The process of inserting a catheter into a blood vessel requires physicians to know the exact location and orientation of the catheter’s distal tip at all times. For example, during a coronary catheterization, the physician must be able to follow the movement of a catheter through a patient’s artery until the affected site is identified [192]. During another critical procedure, percutaneous transluminal coronary angioplasty (PTCA), the physician should accurately position the deflated balloon of a dilatation catheter in an artery before expanding it in order to compress an obstruction [193].  50  To reduce complications, catheterizations are typically performed with assistance of an imaging technique. For instance, it is common to utilize either X-ray imaging [194-197] or ultrasound [198-201], though MRI has become increasingly popular in the past few years due to the feasibility of visualizing soft tissue with a high resolution [202-205]. However, X-ray imaging, specifically fluoroscopy, remains the mainstay of diagnostic and interventional catheterizations. Due to their inherent lack of contrast, determining the position of catheters by X-ray imaging can be challenging. Therefore, it has been suggested to incorporate a radiopaque marker to catheters to improve visualization [206].  Despite early efforts to fabricate suitable radiopaque markers in a reproducible way, not many options are currently available. The very first patents were issued in the early 1990s. They all revolved around the use of metal bands around the distal tip of the catheters [207-209]. Today, companies such as ProPlate® offer a variety of options, which include the production of a metal coating by electroplating gold, nickel, silver, rhodium, palladium, tin, and copper onto the surface of the catheters [210]. Furthermore, tungsten-filled polymer bands have also been fabricated and are commercialized by Putnam Plastics. Common polymers employed in this case are polyurethanes [211]. All of these options, nonetheless, require expensive manufacturing processes. Also, due to their pre-defined sizes, most of these markers only work for a limited number of catheters.  Thus far, coating catheters with radiopaque polymer-based scaffolds has not been reported. Interestingly, as discussed in the previous section, a vast amount of work has been done to produce scaffolds made of polymers, like PCL. Different from metal bands and metal-filled polymer bands, polymer-based scaffolds are malleable and extremely thin. Hence, they could theoretically be used to coat catheters without significantly affecting their size. Not much is known about coating catheters with polymer-based scaffolds, but a lot can be learned from techniques historically employed to coat powders, pellets, granules, and tablets with polymers. For example, the hot melt coating technique, which is one of the oldest strategies in the development of solid dosage formulations, requires three simple steps: 1) heating the polymer (i.e., the coating material) slightly above its melting temperature, 2) applying it directly onto the surface of a substrate, and 3) cooling it down until it solidifies [212-215].  In the following sections, we describe the production of a rhenium-doped scaffold by electrospinning a solution containing PCL. Drawing inspiration from the hot melt coating 51  technique, vascular and PTCA dilatation catheters were coated by melting the scaffold directly onto a small section of the catheters’ surface.   3.2 Materials and Methods  This study commenced with the optimization of the experimental conditions for electrospinning PCL solutions in chloroform and methanol in a controlled and reproducible way. Next, a rhenium complex was synthesized, codissolved with PCL in the same mixture of solvents, and electrospun to produce a rhenium-doped scaffold made of PCL. Finally, sections of two clinically used catheters were coated with this scaffold and imaged by 𝜇CT.    3.2.1 Optimization of the Electrospinning Technique  To evaluate the effect of the size of the polymer on the properties of the scaffolds, PCL (Sigma Aldrich) with low- and intermediate-number average molecular weight (𝑀𝑛 = 10 and 45 kDa, respectively) were used. PCL was dissolved in a mixture of chloroform and methanol (70/30% v/v) at a concentration of 20% w/v. The solutions were heated at 45°C and stirred at 600 rpm for half an hour. They were then let to cool down while keeping them at 600 rpm overnight and until the PCL was fully dissolved. A rotational rheometer (MCR 301, Anton Paar; Graz, Austria) was utilized to measure the viscosity of the solutions at 25 °C across a shear rate range between 1 and 7 s-1.  The electrospinning unit (Kato Tech Co., Ltd.; Kyoto, Japan), or simply electrospinner, utilized to produce the scaffolds is shown in Figure 3.2. A total of 8 mL of the PCL solutions were withdrawn into a 10 mL syringe and electrospun through a 20G 1/2’’ blunt needle. The electrospinning distance was set to 20 cm, the distance advancement rate to 0.07 mm min-1, and the voltage to 25 kV. The system was maintained in a horizontal position. The collecting plate was wrapped in aluminum foil, which acted as conducting material, and it was rotated at a speed of 20 cm min-1. The conducting material promotes the deposition of the NFs, which are electrically charged. To force the NFs to accumulate in a well-defined area and minimize waste, strategic areas of the aluminum foil were covered with paper, which acted as insulating material. Figure 3.3 shows all the different configurations that were tested.   52    Figure 3.2. Electrospinning Setup. (A) The electrospinner is located inside a chamber made of glass with a panel to adjust the settings (B) The image shows (a) the collecting plate and (b) the syringe holder. Currently, this system is commercialized by Kato Tech Co., Ltd. (Kyoto, Japan). Location: Faculty of Pharmaceutical Sciences at UBC in Vancouver, Canada.     Figure 3.3. Configurations for the Collection of Scaffolds. The scaffolds were collected in rectangular strips with different widths, between 2 and 6 cm. The gap between the strips has a width of 4 cm in 3.4.D and 2 cm in 3.4.E. The dark grey areas represent the conducting material (aluminum foil) whereas the light grey areas represent the insulating material (paper).   The experiments were conducted at a temperature of 25 °C and a relative humidity of 40% while keeping the electrospinner’s chamber fully closed. The electrospinning time was ~12 h in all cases. A summary of the scaffolds produced is included in Table 3.1.  53  Table 3.1. Settings for Electrospinning of PCL-Based Scaffolds. As per indicated in Figure 3.3, various configurations were tested to collect the scaffolds. Settings such as electrospinning distance (20 cm), flow rate (0.07 mm min-1), voltage (25 kV), and rotational speed of the collecting plate (20 cm min-1) were kept constant across the experiments. The concentration of PCL in all the solutions was 20% w/v.   Scaffold  ID Number Average Molecular Weight  of the Polymer (kDa)  Configuration  Type Configuration Dimensions [Length x Width]  (cm) S01 10 Figure 3.4B 31 x 4 S02 45 Figure 3.4A 31 x 2 S03 45 Figure 3.4B 29 x 4 S04 45 Figure 3.4C 31 x 6 S05 45 Figure 3.4D 31 x 6 | 31 x 2 S06 45 Figure 3.4E 31 x 2 | 31 x 4 | 31 x 2   The thicknesses of the scaffolds were measured by brightfield microscopy using an inverted microscope (AE31, Motic; Kowloon, China) equipped with a digital camera (Moticam 3, Motic; Kowloon, China). To do this, 0.25 cm2 square-shaped samples of the scaffolds were cut and mounted vertically on a microscope slide. To keep them in a vertical position, the samples were placed between two thin metallic rods. Using a 4x objective, images of different sections of the samples were acquired. ImageJ was then used to measure the thicknesses of the samples, as depicted in the micrographs. For each sample, the thickness was reported as the mean value of several independent measurements on different sections of the samples.  To investigate their morphology at the nanoscale, the scaffolds were imaged by scanning electron microscopy (SEM; SU3500, Hitachi; Chiyoda, Japan). Following the strategy followed before, 0.25 cm2 square-shaped samples of the scaffolds were cut. The samples were mounted onto a specimen stub covered with an electrically conductive carbon-based adhesive disc. They were then sputter-coated with an 8 nm layer of iridium using a modular high vacuum coating system (EM MED020, Leica; Wetzlar, Germany) under reduced pressure (<5 Pa). The images were acquired at 15 kV with a magnification of 10,000x.   3.2.2 Development of a Rhenium-Doped Scaffold by Electrospinning  The next two sections describe the steps followed to produce and characterize a PCL scaffold doped with a rhenium complex, emphasizing the techniques employed to assess the scaffold’s thickness, morphology of the NF network, and cytotoxicity.   54  3.2.2.1 Production and Characterization of the Scaffold  A phosphinophenolate ligand (5-methyl-2-phenoxydiphenylphospine, MeHPO) was synthesized followed by rhenium coordination [216,217]. Both MeHPO and the ensuing rhenium phosphinophenolate complex [ReOCl(MePO)2] were characterized using various spectroscopic techniques: nuclear magnetic resonance (NMR), infrared spectroscopy (IR), and electrospray ionization mass spectroscopy (ESI-MS). MeHPO: 31P{1H} NMR (CDCl3, 202.5 MHz) δ: 25.77 (s). ReOCl(MePO)2: IR (cm-1): 957 (Re=O), ESI-MS: m/z = 785 ([M - Cl]+), 31P{1H} NMR (CDCl3, 202.5 MHz) δ: 15.25, 2.05. For electrospinning, ReOCl(MePO)2 was dissolved at a concentration of 2% w/v in a mixture of chloroform and methanol (70/30% v/v). Subsequently, 45 kDa PCL (Sigma Aldrich) was added at a concentration of 20% w/v and 2.5 mL of the solution was electrospun following the procedure described in Section 3.2.1. This rhenium-doped scaffold was collected on a 3 cm strip for ~4 h. A rhenium-free scaffold (i.e., without the rhenium complex) was produced under the same experimental conditions.  For both scaffolds, the thickness and morphology were assessed by brightfield microscopy and SEM as described in the previous section.   3.2.2.2 Appraisal of the Cytotoxicity  The toxicity of the rhenium-doped scaffold was evaluated in vitro using HEK-293 cells (ATCC® CRL-1573TM; Manassas, USA), a cell line derived from human embryonic kidney cells [218]. The cells were cultured at 37 °C and 5% CO2 in Dulbecco’s Modified Eagle Medium (DMEM; ThermoFisher Scientific) supplemented with 10% w/v fetal bovine serum (FBS; ThermoFisher Scientific) and 1% w/v penicillin-streptomycin, or simply Pen-Strep (10,000 U mL-1, ThermoFisher Scientific). This study consisted of two stages: 1) incubation of the samples in the same media used to culture the cells and 2) real-time cytotoxicity assay using the IncuCyte® Zoom platform (Essen BioScience; Michigan, USA).  Square-shaped samples of the scaffold were cut and incubated in vials with 3 mL of DMEM supplemented with 10% w/v FBS and 1% w/v Pen-Strep. The samples were folded over many times in order to achieve the following concentrations of rhenium: 0.2 mM (one layer, 1x), 0.4 mM (two layers, 2x), 0.8 mM (four layers, 4x), and 3.2 mM (sixteen layers, 16x). The incubation was carried out in a benchtop shaker (SI-300, Lab Companion; Seoul, Korea) at 37 °C 55  and 150 rpm. After 24 h, the samples were removed from the vials and the media were sterile filtered using a syringe filter unit with a 0.22 𝜇m pore size (Millex®). Furthermore, samples of the rhenium-free scaffold were cut and incubated under the same experimental conditions. As per indicated in Table 3.2, the weights of pieces with the same number of layers were maintained consistent regardless of the composition of the scaffolds (i.e., rhenium-doped or rhenium-free).   Table 3.2. Incubation of Samples of Scaffolds for Cytotoxicity Assay. A 24 h incubation was carried out in DMEM supplemented with 10% w/v FBS and 1% w/v Pen-Strep. Different concentrations of rhenium were achieved by using samples made of one, two, four, or sixteen layers of the rhenium-doped scaffold.   Scaffold  ID Scaffold  Type Number  of Layers Scaffold’s Mass Concentration (mg mL-1) Rhenium’s Molar  Concentration (mM) S07 Rhenium-Doped  Scaffold 1x 0.4 0.2 2x 0.8 0.4 4x 1.6 0.8 16x 6.4 3.2 S08 Rhenium-Free Scaffold 1x 0.4 N/A 2x 0.8 N/A 4x 1.6 N/A 16x 6.4 N/A   The IncuCyte® Zoom monitors changes in confluency over time (i.e., an estimate of the growth rate of the cells). The confluency is the proportion of the surface of the wells which is covered by cells (e.g., a confluency of 25% means that the cells occupy a quarter of the surface and thus still have plenty of room to grow). Therefore, the confluency is a measure of the number of cells per well [219,220]. The fact that the IncuCyte® Zooms monitors the growth rate of the cells in real time differs from traditional cytotoxicity assays, where only an end-point cell death marker can be evaluated: the release of a compound from cells with a compromised membrane (i.e., dead cells) at a single time point (usually 24 h post-treatment). For example, two commonly used colorimetric assays, MTT (named after 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoium bromide, a tetrazolium dye) and LDH (named after lactate dehydrogenase, an enzyme), measure the release of formazan and lactate dehydrogenase, respectively. None of these assays provide information about changes in the growth rate of the cells over time [221,222]. The more advanced IncuCyte® Zoom continuously monitors changes in the cells after treatment and even acquires images of the cells over the entire duration of the experiment, thereby providing more accurate and reliable information [219,220].  56  To start the assay with the IncuCyte® Zoom, 5,000 cells per well were seeded in a 96-well plate. After 24 h, the media were removed and the wells were replenished with 100 𝜇L of fresh media. Then, 100 𝜇L of each sample were added per well. A total of eight wells were allocated per sample (i.e., eight replicates). As a blank and vehicle control, 100 𝜇L of media were added to all the remaining wells. The change in confluency during the first 80 h post-treatment was measured using the IncuCyte® Zoom software, developed by the incubator’s manufacturer.  Leakage, or release, of ReOCl(MePO)2 from the scaffolds’ matrix was investigated by inductively coupled plasma mass spectroscopy (ICP-MS). This technique was used to detect traces of rhenium in the media used to incubate the samples of the rhenium-doped scaffold. As controls, the following samples were also analyzed: 1) the media utilized to incubate the samples of the rhenium-free scaffold and 2) a series of ammonium perrhenate solutions with a known concentration of rhenium (0.2, 10, and 100 mM). To prepare the samples, 200 𝜇L of each sample were mixed with 8 𝜇L of concentrated hydrochloric acid (HCl) and the solutions were filled up to 2 mL with water. The samples were sent to Exova in Surrey, Canada to complete the analysis.  As explained at the end of Section 1.2, rhenium complexes oxidize to perrhenate in vivo. In the event of leakage of ReOCl(MePO)2 from the scaffolds’ matrix in vivo, this transformation is thus expected to occur. Therefore, assessing the toxicity of perrhenate is very important. The IncuCyte® Zoom was employed to evaluate the toxicity of a series of ammonium perrhenate solutions with a concentration of rhenium between 0.1 and 100 mM. The solution with a concentration of rhenium of 100 mM was prepared by dissolving ammonium perrhenate in 0.9 w/v sodium chloride (NaCl, or simply saline). The rest of the samples were prepared by serial dilutions. For the assay, 10 𝜇L of the samples were added to wells seeded with 5,000 cells in 190 𝜇L of media. The following controls were used: 200 𝜇L of media as a blank and a mixture of 10 𝜇L of saline and 190 𝜇L of media as a vehicle control. A total of eight wells were allocated for each treatment, including controls.   3.2.3 Coating and X-Ray Imaging of Catheters  A similar approach than the hot melt coating technique was used to incorporate a rectangular-shaped sample of the rhenium-doped scaffold onto the surface of a short section of a vascular catheter made of polyethylene (PE-50, ID = 0.05 cm and OD = 0.10 cm, where ID and OD stand for inner and outer diameters, respectively; IntramedicTM, Becton Dickinson). The 57  sample had a length of 1.5 cm and a width of approximately 2 mm but it was actually constituted of eight layers, which facilitated handling and processing. The catheter was wrapped with the sample of scaffold carefully but tightly. The scaffold/catheter system was then completely covered with aluminum foil and placed directly on the surface of a hot plate equipped with a temperature probe. The temperature was initially set to 55 °C but it was slowly increased in 1 °C increments until the scaffold melted, which occurred at 59 °C. The catheter was removed from the hot plate and allowed to cool down. The same procedure was utilized to coat another section of the same catheter with the rhenium-free scaffold.   Following the same steps, the balloon in a PTCA dilatation catheter (Long Cobra 40TM, SciMed Life Systems, Inc.) was coated with the rhenium-doped and the rhenium-free scaffolds. The scaffolds were applied directly on top of the balloon, without air (i.e., not expanded).  The coated sections of the catheters were cut and placed inside microcentrifuge tubes for imaging by 𝜇CT using a high-resolution specimen scanner (𝜇CT 100, SCANCO Medical; Brüttisellen, Switzerland) (Figure 3.4). The samples were imaged at 90 kVp (200 𝜇A and 550 ms). According to the manufacturer’s specifications, the only source of filtration in the scanner is the X-ray window, made of 0.15 mm of beryllium. The acquisition consisted of a 1,000 projections. Using MicroView (Parallax Innovations; Ilderton, Canada), the data was reconstructed as a 3D image with a matrix size of 2048 x 2048 and a spatial resolution a slice thickness of 5 𝜇m.    Figure 3.4. Specimen Micro-Computed Tomography Scanner. (A) The instrument can operate at up to 90 kVp with a resolution between 1.25 and 200 𝜇m. The equipment is fully shielded. (B) The instrument has a chamber which can accommodate up to twelve sample holders, allowing the automatization of large studies. After loading, the sample holders are imaged consecutively. Model and Manufacturer: 𝜇CT 100, SCANCO Medical (Brüttisellen, Switzerland). Location: CHTP at UBC in Vancouver, Canada.    58  3.3 Results and Discussion  Our study started with the optimization of the electrospinning technique in order to produce scaffolds constituted of PCL. After optimization of the experimental conditions (i.e., concentration and molecular weight of PCL, volume of the solution, configuration for scaffold collecting, electrospinning distance, distance advancement rate, and voltage), we produced a radiopaque electrospun scaffold by the incorporation of a rhenium complex into the scaffold’s matrix. After coating them with samples of this scaffold, a couple of otherwise radiolucent catheters became visible in images acquired by 𝜇CT. A discussion of the experimental findings derived from this work is provided below.   3.3.1 Optimization of the Electrospinning Technique  The optimization of the electrospinning technique aimed: 1) to produce scaffolds in a controlled and reproducible way and 2) to collect scaffolds with adequate physical properties, as dictated, for example, by their ease of handling. To accomplish this, the scaffolds’ structure at both the macroscale and the nanoscale was investigated in detail.  Representative images of the scaffolds prepared with 45 kDa PCL are depicted in Figure 3.5 right after collection. Furthermore, their thicknesses are summarized in Table 3.3. The thickness of the scaffold fabricated with 10 kDa PCL could not be measured due to difficulty to handle and manipulate the scaffold, which was very thin and brittle. Deposition of NFs on the insulating material was quite significant in configurations with only one strip. However, scaffolds with roughly the same thickness were produced in all these cases. These results suggest that there is a limited amount of NFs which can be deposited on the strip before they start to be deposited on the insulating material. Due to the strong adherence of the scaffold to the insulating material, NFs deposited outside the strip are considered waste. Therefore, efforts should be made to maximize the deposition of NFs on the strip, or area of interest.   59    Figure 3.5. Examples of 45 kDa Polycaprolactone-Based Scaffolds. (A-E) The pictures show the scaffolds before removal of the insulating material. To improve their visibility, the periphery of the scaffolds was delineated in black.  A high amount of NFs were deposited outside the strip in configurations (A), (B), and (C). Based on Table 3.1, the scaffolds are: (A) S02, (B) S03, (C) S04, (D) S05, and (E) S06. (F) The picture depicts the two scaffolds shown in (D) after removal of the insulating material. Due to its increased thickness, the primary scaffold (on the left) is more visible than the secondary scaffold (on the right).    Table 3.3. Thickness Appraisal of 45 kDa Polycaprolactone-Based Scaffolds. The values were attained by measuring the thicknesses of end and middle sections of the scaffolds in images acquired by brightfield microscopy. For S05 and S06, the thicknesses of the scaffolds formed in each strip are reported.    Scaffold  ID Thickness (𝝁m)  S02 910.9 ± 35.2 S03 931.8 ± 69.7 S04 886.5 ± 34.9 S05 958.7 ± 115.8 | 538.4 ± 59.3 S06 710.8 ± 23.6 | 783.1 ± 57.6 | 453.1 ± 56.9   Deposition of NFs in configurations with more than one strip occurred preferentially on the strip located directly opposite to the syringe (referred herein as primary scaffold production). The strips located on the right or left sides of the central strip were filled with NFs only after the first 6-8 h of collection (referred herein as secondary scaffold production). These findings suggest that the primary scaffold starts to behave as an insulating material after reaching a specific thickness. At this point, deposition of NFs on the side strips, or on the insulating material in 60  configurations with only one strip, is therefore favored. This phenomenon is attributable to charge buildup within the primary scaffold’s matrix.  As the NFs accumulate on the strip, the electric field at the tip of the Taylor cone is reduced due to residual charges in the scaffold. Usually, residual charges are dissipated by mechanisms depending upon a number of parameters, such as distribution of charges within the scaffold’s matrix, dielectric properties of the polymer, temperature, and relative humidity [223]. Interestingly, charge dissipation is often not complete and, with the use of electrostatic fieldmeters, a measurable amount of residual charges have in fact been found on electrospun scaffolds [224]. For example, in semicrystalline polymers, like PCL, charge buildup is believed to occur at the interface of crystalline and amorphous regions [225]. Generally, charge buildup can have two consequences: 1) diversion of the jet to a more conductive region or 2) termination of the process due to reduction of the electric field below a threshold required for electrospinning. The use of additional strips facilitates the redirection of the jet to specific locations when the charge density in the primary scaffold becomes too high.  The configuration with two strips resulted in a primary scaffold with similar thickness than the scaffolds produced in configurations with only one strip. The configuration with three strips resulted in a primary scaffold with reduced thickness. However, it also led to minimal deposition of NFs on the insulating material. An important remark to make is that the secondary scaffolds were not produced simultaneously. For instance, the secondary scaffold located on the left side of the central strip was formed first and it ultimately reached a thickness similar to the primary scaffold. Therefore, the slightly higher thicknesses of the scaffolds produced in all other configurations are deemed to be caused by non-specific deposition of NFs on the primary scaffold after charge buildup. To put it another way, the use of the configuration with three strips showed that once the primary scaffold reached a thickness of around ~780 𝜇m, deposition of NFs on that strip ceased and continued on a different strip until a similar thickness was achieved.  These results indicate that increasing the volume of the solution and the electrospinning time does not necessarily produce thicker scaffolds but it might increase the deposition of NFs on the insulating material. Thus, reducing the volume of the solution and collecting the scaffold on a single, well-defined strip could be really helpful in achieving a desired thickness while minimizing the amount of waste. This is of particular importance when working with materials available only in limited quantities. In fact, this is the case of ReOCl(MePO)2, the rhenium 61  complex synthesized for incorporation into a scaffold made of PCL (i.e., our next series of experiments).  PCL has been extensively used to prepare scaffolds by electrospinning, but not much is known of the effect of its molecular weight on the physical properties of the scaffolds [226,227]. The molecular weight influences the rheological properties of the solution, like the viscosity. The effect of the molecular weight on the viscosity is related to the molecular entanglement of the polymer chains [228]. Briefly, the molecular entanglement model refers to polymer chains as a network of bridges which are long enough to form at least one loop on themselves: the higher the molecular weight, the stronger the bridges. At the same concentration, solutions prepared with polymers with a higher molecular weight are thus significantly more viscous [229].  The viscosity is regarded as the most critical property governing the production of scaffolds by electrospinning and it has a great influence on the morphology of the NF network. Particularly, when really low-viscosity solutions are electrospun, continuous and smooth NFs cannot be produced. As a matter of fact, nanobeads are oftentimes produced in addition to NFs in these cases [230]. As discussed in Section 3.1.1, the formation of NFs occurs only after the jet undergoes a bending instability, which depends on the viscosity of the solution. Below a viscosity threshold, the jet undergoes a Rayleigh instability instead of a bending instability, which results in the breakup of the jet into droplets [231]. Conversely, when really high-viscosity solutions are electrospun, the jet cannot be ejected from the capillary at a steady rate, resulting in the formation of irregular NFs, usually larger and thicker [228].   Figure 3.6 depicts the variation in the viscosity as a function of the shear stress for the two solutions prepared with 10 and 45 kDa PCL in a mixture of chloroform and methanol (70/30% v/v) at a concentration of 20% w/v. The curves show that the higher the shear rate, the larger the viscosity. This behavior is characteristic of shear thickening or dilatant fluids, a class of non-Newtonian fluids. However, the mixture of solvents is known to be a Newtonian fluid [232], in other words, its viscosity is independent from the shear rate. The non-Newtonian behavior of the polymer-based solutions is therefore due to the presence of the polymer in the solution. The graph also shows that the viscosity of the solution prepared with 45 kDa PCL is more than 10 times higher than the viscosity of the solution prepared with 10 kDa PCL. This in turn explains why the scaffold prepared with 10 kDa PCL was very brittle and thus lacked any practical applications, whereas all the scaffolds prepared with 45 kDa PCL were easy to cut and fold.  62   Figure 3.6. Viscosity of Polycaprolactone-Based Solutions. The solutions prepared with 10 and 45 kDa PCL exhibit a non-Newtonian behavior. The solution containing 45 kDa PCL has a significantly higher viscosity than the one containing 10 kDa PCL as a result of a higher degree of molecular entanglements.   Upon inspecting the SEM images, it became evident that the scaffolds’ structure at the macroscale correlates with their morphology at the nanoscale. Figure 3.7A shows that the 10 kDa PCL-based scaffold was constituted of very thin NFs interconnected through nanobeads and other amorphous structures. Figures 3.7B-F depict representative images of the 45 kDa PCL-based scaffolds. As expected, no differences were observed in the morphology of any of these scaffolds. Indeed, all the scaffolds fabricated with 45 kDa PCL were characterized by the presence of a smooth and continuous network of non-porous NFs with minimal or no presence of nanobeads, opposite to what was observed in scaffolds produced with 10 kDa PCL.   63    Figure 3.7. Morphology Inspection of Polycaprolactone-Based Scaffolds. Images acquired by SEM are shown for samples of (A) the 10 kDa PCL-based scaffold and (B-F) the 45 kDa PCL-based scaffolds. Based on Table 3.1, the scaffolds are: (A) S01, (B) S02, (C) S03, (D) S04, (E) S05, and (F) S06.    Our study has provided a better understanding of important properties which influence the structure of scaffolds produced by electrospinning. For example, it was shown that the viscosity has an effect on the scaffolds’ ease of handling, thickness, and morphology of the NF network. Furthermore, the viscosity influences the choice of certain operational parameters, such as distance advancement rate and voltage. Although other properties like size (e.g., diameter, surface area-to-volume ratio) and architecture (e.g., porosity, anisotropy, tortuosity) of the NFs are also descriptive of the physical properties of the scaffolds, they are usually overlooked due to the lack of access to appropriate instrumentation [233].  The viscosity depends on both the concentration of the polymer in the solution and the molecular weight of the polymer. Herein, only one concentration was evaluated: 20% w/v. This is the minimum concentration that has been reported to produce electrospun scaffolds with a smooth and continuous network of NFs using high-molecular weight PCL: 65-67 [234-236] and 80 [237-241] kDa. Generally, it has been observed that nanobeads form in addition to NFs at lower concentrations of typically 8 to 15% w/v [242-246]. Considering that this phenomenon is attributed to a reduction in the viscosity, 20% w/v is the minimum concentration which could theoretically allow the preparation of scaffolds constituted of NFs alone when using PCL with a 64  molecular weight below 80 kDa. Regarding the molecular weight, two options were tested: low-molecular weight (i.e., 10 kDa) and intermediate-molecular weight (i.e., 45 kDa). Our study demonstrated that 10 kDa PCL at a concentration of 20% w/v forms a brittle scaffold, very difficult to handle. However, at the same concentration, we successfully showed that 45 kDa PCL can be utilized to produce a scaffold with better physical properties.   3.3.2 Development of a Rhenium-Doped Scaffold by Electrospinning  The results discussed in the previous section indicated that the next stage of this study had to begin with the preparation of a polymer-based solution using 45 kDa PCL at a concentration of 20% w/v. As before, PCL was added to a mixture of chloroform and methanol (70/30% v/v), but adding at the same time a rhenium complex before electrospinning. The following section describes the results derived from this work, including a discussion of both the physical properties and the cytotoxicity of the newly produced rhenium-doped scaffold.    3.3.2.1 Production and Characterization of the Scaffold  The rhenium phosphinophenolate complex synthesized for incorporation into a PCL-based scaffold is a stable, neutral octahedral complex with a P2O3Cl coordination sphere. The structure of this lipophilic rhenium complex is shown in Figure 3.8, where it can be observed that the MePO- ligands work as bidentate chelators. A chloride fills the last coordination site around the Re=O core [217]. This rhenium complex is a dark green powder and is miscible in organic solvents, including chloroform. This made it an ideal candidate for incorporation into a solution of 45 kDa PCL in 70/30% v/v chloroform/methanol (successfully used before to produce electrospun scaffolds). The molecular weight of ReOCl(MePO)2 is 820.3 Da. Considering that the atomic mass of rhenium is 186.2 Da, the mole percent rhenium content in ReOCl(MePO)2 was found to be 22.7%, calculated as follows: (186.2 Da / 820.3 Da) (100).    65    Figure 3.8. Rhenium Phosphinophenolate Complex. The image shows the ball-and-stick model of ReOCl(MePO)2, where Re, Cl, O, and P are depicted in blue, green, red, and orange, respectively. The crystal structure of this rhenium complex was reported by Kovacs et al. (Cambridge Structural Database: 168621) [217].  For X-ray imaging, the source of contrast in the scaffold is rhenium. The mole percent rhenium concentration in the scaffold was estimated to be 19.2%. This value was calculated by dividing the number of moles of rhenium in the scaffold by the total number of moles in the scaffold (i.e., number of moles in 0.05 g of ReOCl(MePO)2 and 0.5 g of PCL). The quotient was multiplied by 100 to express the result as a percentage. The number of moles of rhenium was calculated by multiplying the number of moles of ReOCl(MePO)2 by the mole percent rhenium content in ReOCl(MePO)2 as follows: (0.05 g / 820.3 g mol-1) (0.227). The total number of moles was calculated as follows: (0.05 g / 820. 3 g mol-1) + (0.5 g / 45,000 g mol-1).  To maximize the deposition of NFs on the area of interest (i.e., a 3 cm wide strip covered with aluminum foil), the volume of the rhenium-doped polymer-based solution was reduced from 8 to 2.5 mL. As discussed in Section 3.3.1, there is a limited amount of NFs which can be deposited in a preferential way on the strip. The electrospinning time was adjusted accordingly, from ~12 to ~4 h. Despite noticing the deposition of NFs outside the strip during collection, only extremely thin scaffolds were formed on the insulating material. Therefore, it is thought that the majority of the NFs were in fact deposited on the strip. The same was noticed for the rhenium-free scaffold (i.e., without the rhenium complex). As expected, with the only exception that the rhenium-doped scaffold was green and the rhenium-free scaffold was white, the scaffolds were virtually identical. The light green tint of the rhenium-doped scaffold is a result of blending the rhenium complex (which is dark green) with the polymer (which is white). However, both scaffolds were equally easy to handle, cut, and fold. The thicknesses were found to be 779.9 ± 127.1 𝜇m for the rhenium-doped scaffold and 722.4 ± 96.1 𝜇m for the rhenium-free scaffold. These values are in agreement with the results attained 66  before at the same experimental conditions: the deposition of NFs on a single strip occurs preferentially until the scaffold reaches a thickness of around ~780 𝜇m (from Section 3.3.1).  The incorporation of ReOCl(MePO)2 into the scaffold’s matrix did not alter the morphological features of the NFs network. As shown in Figure 3.9, both scaffolds, rhenium-doped and rhenium-free, were found to be made of a smooth and continuous non-porous NF network indistinguishable from each other. Similar to all the scaffolds produced before with 45 kDa PCL, minimal nanobeads formation was observed in these two scaffolds.      Figure 3.9. Morphology Inspection of Rhenium-Doped and Rhenium-Free Scaffolds. Images acquired by SEM are shown for (A) the rhenium-doped scaffold and (B) the rhenium-free scaffold. A pseudocolor was assigned to (A) to reflect the tint observed in the rhenium-doped scaffold.   3.3.2.2 Appraisal of the Cytotoxicity  A potential source of toxicity associated with the rhenium-doped scaffold is leakage of ReOCl(MePO)2 from the scaffold’s matrix. As discussed in Section 1.2, rhenium complexes oxidize in vivo to perrhenate upon administration into the body [8,26]. Perrhenate is naturally cleared by the kidneys [27-30]. Thus, herein we chose to evaluate the toxicity of the rhenium-doped scaffold in vitro in human embryonic kidney cells, specifically HEK-293 cells [218].  Figure 3.10 depicts the change in confluency during the first 80 h post-treatment. No difference was found between the rhenium-doped and the rhenium-free scaffolds relative to the vehicle control. All the curves reach a plateau at around 72 h post-treatment, where the confluency is between 90 and 100%. The small variation between the curves is attributable to the initial difference in confluency. For instance, the graph shows that the difference in confluency between the curves 80 h post-treatment is around 10%, which is roughly the same than at the beginning of the experiment.  67    Figure 3.10. HEK-293 Cell Confluency Variation Post-Scaffold Treatment. The cells were treated with samples of media in which pieces of the scaffolds were incubated at 37 °C for 24 h. For the rhenium-doped scaffold, the following concentrations of rhenium were evaluated: 0.2 mM (one layer, 1x), 0.4 mM (two layers, 2x), 0.8 mM (four layers, 4x), and 3.2 mM (sixteen layers, 16x).     Through ICP-MS, it was found that no more than 0.3% of the total amount of rhenium in the rhenium-doped scaffold was released from the polymer-based matrix after a 24 h long incubation. For example, 0.6 𝜇M of rhenium was detected in the sample with 0.2 mM of rhenium (one layer, 1x) and 8.1 𝜇M of rhenium was detected in the sample with 3.2 mM of rhenium (sixteen layers, 16x). This in turn explains why no difference was observed between cells treated with the rhenium-doped scaffold and those treated with the rhenium-free scaffold.  Due to the transformation of rhenium complexes to perrhenate in vivo, the toxicity of perrhenate is of great interest. In our study, ammonium perrhenate was not found to be toxic at a concentration as high as 100 mM (Figure 3.11), which is consistent with other studies [247,248]. Therefore, toxicity associated with the presence of perrhenate is of no concern even if a higher amount of rhenium is released from the scaffold’s matrix. This situation is only expected to happen if the scaffold is placed inside the body for a prolonged period of time.   68    Figure 3.11. HEK-293 Cell Confluency Variation Post-Ammonium Perrhenate Treatment. The cells were treated with samples of solutions of ammonium perrhenate in saline with a concentration between (A) 0.1 and 1.6 mM and (B) 3.1 and 100 mM.   3.3.3 Coating and X-Ray Imaging of Catheters  Figures 3.12A and C show images of the two catheters after coating them with samples of the rhenium-doped and the rhenium-free scaffolds. This study showed that a uniform coating is achieved when the temperature is increased slowly, specifically in increments of 1 °C. Increasing the temperature rapidly to 59 °C resulted in some of the material adhering to the aluminum foil instead of the catheter. As mentioned in Section 3.2.3, the scaffold/catheter system was covered with aluminum foil during the melting step. Interestingly, the scaffolds did not adhere to the catheters when aluminum foil was not used. Due to its highly reflective surface, aluminum foil reduces the loss of heat. This might be helpful in ensuring an even distribution of heat across the scaffold/catheter system, thereby allowing the controlled deposition of the scaffolds onto the catheters. Figures 3.12B and D depict representative 𝜇CT axial images of the catheters, showing that visualization of the catheters was significantly enhanced with the rhenium-doped polymer coating. The images show that the rhenium-doped polymer-based coating has a higher contrast than the rhenium-free polymer-based coating. This was achieved by the addition of ReOCl(MePO)2 into the scaffold’s matrix, with a mole percent rhenium concentration of 19.2% (from Section 3.3.2.1). Moreover, no variation in the contrast was observed across the rhenium-doped polymer-based coating. The coating was produced by melting (and thus, destroying) the NFs, but the NFs might actually be important in achieving a homogenous distribution of ReOCl(MePO)2 within the scaffold’s matrix, resulting in an uniform, strongly radiopaque 69  coating. Due to the large surface area-to-volume ratio of the NFs, it has been reported that compounds are usually homogenously distributed in electrospun scaffolds [249]. Exceptions occur when the compounds are not soluble in the polymer-based solution and they start to precipitate in the syringe during electrospinning. In these cases, the compounds migrate to the surface of the NFs [250].     Figure 3.12. X-Ray Imaging of Coated Catheters. The images show (A, B) the vascular catheters and (C, D) the PTCA dilatation catheters coated with samples of (a) the rhenium-doped scaffold and (b) the rhenium-free scaffold. (B, D) Representative axial images acquired at 90 kVp (110 𝜇As). (B) corresponds to the catheters in (A) and (D) corresponds to the catheters in (C). A significant improvement in contrast is observed with the rhenium-doped polymer-based coating. Four uncoated catheters are included in (B) (indicated with red arrows).   Poor distribution of ReOCl(MePO)2 would have resulted in localized regions of high radiopacity in an otherwise radiolucent coating, which is not the case of the rhenium-doped polymer-based coatings shown in Figures 3.12B and D. This was quantitatively demonstrated by determining the distribution of grey values across the coatings. This analysis was performed using Image-Pro Plus 7 (Media Cybernetics; Rockville, USA), an image analysis software. A circle was drawn inside the coatings of both catheters in the 𝜇CT axial images shown in Figures 3.12B and D. The intensities of all the pixels in the circles were then calculated in grey values. This analysis was also performed in uncoated sections of the catheters.   70  The variation in the pixel intensities of the coatings is shown graphically in Figure 3.13A for the vascular catheter and Figure 3.13E for the PTCA dilatation catheter. In the same image, 3D plots of the distribution of grey values in the coatings are also shown, created with the same software. The mean pixel intensities calculated for the coatings are reported in Table 3.4. For the vascular catheter, there is a 142% increase in pixel intensity when the catheter is coated with the rhenium-doped polymer-based coating (calculated relative to the pixel intensity of an uncoated section of the catheter). For the PTCA dilatation catheter, there is a 109% increase in pixel intensity (calculated relative to the pixel intensity of the balloon, without a coating). Furthermore, as suggested by the standard deviations, which fall within 10% of the mean values, there is minimal variation in the pixel intensities of the coatings. This analysis thus corroborates that the rhenium-doped polymer-based coating shows a very uniform contrast.    Figure 3.13. Pixel Intensity Analysis of the Coatings. The pixel intensities were determined by measuring the grey values across circles drawn inside the coatings using Image-Pro Plus 7 (Media Cybernetics; Rockville, USA). The variation in pixel intensity across the coatings is shown in (A) for the vascular catheter and (B) for the PTCA dilatation catheter. 3D plots of the distribution of grey values are shown for the vascular catheter in (B), (C), and (D) and for the PTCA dilatation catheter in (F) and (G). Samples: (B) and (F) rhenium-doped polymer-based coating, (C) and (G) rhenium-free polymer-based coating, and (D) no coating.  71  Table 3.4. Mean Pixel Intensity of the Coatings. The mean pixel intensity of each structure was calculated by taking the average of all the grey values (see Figure 3.13). For the uncoated sections, the analysis was performed on the plastic for the vascular catheter and on the balloon for the PTCA dilatation catheter.   Structure Mean Pixel Intensity Vascular Catheter PTCA Dilatation Catheter Rhenium-Doped Polymer-Based Coating 75.5 ± 7.1 39.5 ± 5.0 Rhenium-Free Polymer-Based Coating 49.5 ± 5.2 25.4 ± 3.2 No Coating 31.3 ± 5.0 18.9 ± 2.3   3.4 Conclusions  A biodegradable and biocompatible electrospun rhenium-doped scaffold made of PCL was successfully produced and characterized in terms of its thickness, morphology, cytotoxicity, and radiopacity. Using a similar approach than the hot melt coating technique, we showed that samples of the scaffold can be adhered to the surface of clinically utilized catheters. After X-ray imaging, the sections of the catheters coated with the rhenium-doped scaffold displayed a substantial improvement in contrast.   72  Chapter 4 Development and Imaging of Radiopaque and Radioactive Microspheres  Thorough experimental evidence of the X-ray attenuation of rhenium was provided in the previous two chapters. An application of rhenium in X-ray imaging was specifically discussed in Chapter 3, where a strongly radiopaque coating for catheters was fabricated by melting an electrospun rhenium-doped scaffold directly onto their surface. This rhenium-doped polymer-based coating successfully turned the catheters visible in images acquired by 𝜇CT. Not only large medical devices, but also microsized carriers, such as MS, can be made radiopaque by incorporation of rhenium.  Embolization therapy, or embolotherapy, encompasses all various procedures where MS are injected into patients in order to reduce or fully impede blood flow to an area of the body. Embolotherapy is a particularly common treatment option in patients with liver cancer which cannot be removed by surgery. The administration of MS is performed under X-ray imaging guidance using catheters. Since currently available embolic MS are radiolucent, they are typically blended with potentially toxic I-XCAs to visualize them under X-rays. To address this limitation, this chapter revolves around the production of intrinsically radiopaque MS by incorporating rhenium into polymer-based MS.  A common form of embolotherapy in patients with liver cancer is radioembolization. In this procedure, MS loaded with a radioisotope emitting short-range 𝛽− particles are directly administered into the hepatic artery, which provides more than 80% of the blood supply to the tumors. Although 𝛽− particles have a high cell killing power, they cannot be directly imaged. To overcome this problem, this chapter also revolves around the production of “imageable” radioactive MS by labeling polymer-based MS with 188Re, a mixed 𝛽− and 𝛾 emitter. Due to the 𝛾 emissions of 188Re, these MS can be imaged in vivo by SPECT.   4.1 Background  This section includes: 1) a review of the current state of the art in embolotherapy, specifically in the management of primary liver cancer, and 2) a description of the flow focusing method, a microfluidic technology for the fast production of uniformly-sized MS. This method was used for the production of the following two samples of MS made of biodegradable and 73  biocompatible polymers: radiopaque MS doped with a non-radioactive rhenium complex and radioactive MS labeled with a 188Re complex.  4.1.1 Embolization Therapy  Embolotherapy refers to procedures where MS are administered into patients via X-ray imaging-guided catheterizations. These procedures are performed for a number of reasons; for example, to control and/or prevent bleeding, eliminate abnormal connections between arteries and veins, treat aneurysms, and block the blood supply to tumors [251]. Over the past 40 years, various forms of embolotherapy have been extensively used in the treatment of liver, lung, kidney, and cervical cancers [252-255]. Particularly in the management of primary liver cancer, embolotherapy has become the standard of care.   The most common type of primary liver cancer is hepatocellular carcinoma (HCC). In fact, HCC is the sixth most common cancer in the world and the third most common cause of cancer-related death. According to the latest report by the International Agency of Research on Cancer (IARC) in 2012, the incidence of HCC is on the rise with 780,000 new cases, or 5.6% of the total number of new cases of cancer [256]. This malignancy is characterized by poor prognosis with mortality rates close to incidence rates [257]. The 5-year survival rate is less than 5% without treatment [258] Due to delayed detection, tumor morphology (i.e., larger than >5 cm or multicentric), and poor underlying liver function (mostly caused by hepatitis B and C, cirrhosis, and/or malnutrition), only 10% of patients qualify for curative therapies (i.e., ablation, segmental resection, and transplantation) [259-261]. Therefore, the vast majority of patients are subjected to embolotherapy.  Several types of liver embolotherapy are currently utilized in the clinical setting. The state of the art of these procedures is discussed next.   4.1.1.1 Embolization Therapy for Hepatocellular Carcinoma  The liver has a uniquely organized dual blood supply. In HCC, more than 80% of the blood supply to the tumors is derived from branches of the hepatic artery, whereas the blood supply to the healthy liver tissue is predominately derived from the hepatic portal vein [262,263]. Precisely, liver embolotherapy aims to block the branches of the hepatic artery which feed the tumor. To accomplish this, patients undergo a hepatic intra-arterial catheterization where MS are 74  directly administered into the hepatic artery [264]. Depending upon the size and the mechanism of action of the MS, the following procedures fall within the scope of liver embolotherapy: transarterial embolization (TAE), conventional transarterial chemoembolization (cTACE), drug-eluting bead transarterial chemoembolization (DEB-TACE), and radioembolization.  In TAE, 100 to 300 𝜇m MS are administered directly into the hepatic artery to occlude the blood vessels which feed the tumors, ultimately leading to an ischemic and hypoxic environment [264,265]. As the healthy liver tissue receives minimal blood flow from the hepatic artery (<20%), it is spared from these effects [265]. In cTACE, slightly larger MS, 300 to 500 𝜇m in diameter, are co-administered with a high dose of an emulsion of chemotherapeutic agents prepared with an iodine-based ethiodized oil, typically Lipiodol (Laboratoire Andre Guerbet; Aulnay-sous-Bois, France). Like in TAE, the MS employed in cTACE cause ischemia and hypoxia in the tumor but they also contribute to increase the intratumoral retention of the chemotherapeutic agents (doxorubicin alone or in combination with mitomycin C and/or cisplatin) [266,267]. DEB-TACE consists on the administration of drug-loaded 100 to 300 𝜇m MS which provide a sustained drug delivery to the tumor and also cause ischemia and hypoxia. Commonly employed drugs in DEB-TACE are doxorubicin and sorafenib. The adverse effects associated with these drugs (mainly cardiomyopathy and hypertension, respectively) have been significantly reduced as a result of the decreased systemic exposure of the drugs [267-269].  Historically, MS used in TAE, cTACE, and DEB-TACE have been made mostly of trisacryl gelatin and poly(vinyl alcohol) (PVA). These materials, nevertheless, are radiolucent. Prior to injection, the MS are blended with I-XCAs, which act as a surrogate marker of the biodistribution of the MS [270]. The XCAs are administered for three reasons: 1) guide the placement of the catheter in the hepatic artery, 2) monitor unintentional reflux into non-target blood vessels, and 3) corroborate the delivery of the MS into the target tissue [270,271].  The major shortcoming of this approach is that visibility diminishes soon after injection due to I-XCA washout, leaving the final location of the MS unknown. Radiopaque MS may provide procedural value by improving the visualization of target and non-target embolized tissues compared to I-XCAs alone. Also, they will minimize, or even eliminate, the need of co-administering potentially toxic I-XCAs [272-274]. As mentioned in previous sections of this dissertation, I-XCAs have been linked to severe adverse reactions, including anaphylaxis and nephropathy [47-50].   75  Generally, TAE, cTACE, and DEB-TACE represent the mainstay of treatment for patients in whom liver function is preserved. After treatment, the MS stay in the patient’s body indefinitely and may cause vascular injury and abscess formation. This has led to the notion that complete occlusion of large blood vessels may not be an optimal approach in patients with compromised liver function [275]. A fourth type of hepatic embolotherapy, radioembolization, has gained recognition as a viable treatment option for patients considered to be poor candidates for TAE, cTACE, or DEB-TACE [276,277]. In radioembolization, MS functionalized with a radioisotope emitting short-range 𝛽− particles are administered directly into the hepatic artery. The MS, which have a diameter between 20 and 60 𝜇m, lodge near the terminal arterioles of the tumors. As a result of the liver’s compartmentalization of the blood supply, the healthy liver tissue is spared from the radiation, but the tumors are exposed to a tumoricidal radiation dose of 100 to 150 Gy [262,278]. Experience gained in recent years suggests that HCC is truly a radiosensitive tumor [279]. Hence, radioembolization has quickly become the first-line treatment in most HCC patients with early-, intermediate-, and advanced-stage HCC. In some cases, it has been successfully used to downstage patients to ablation, segmental resection, or transplantation [280,281]. The most widely utilized radioembolization agents, and therefore most often compared, are MS made of glass (TheraSphere®, BTG Interventional Medicine; London, United Kingdom) or resin (SIR-Spheres®, Sirtex Medical Limited; New South Wales, Australia) [282]. Both of these MS are loaded with 90Y (a radioisotope of yttrium). These 90Y-MS have, nonetheless, different sizes: 20 to 30 𝜇m for TheraSphere® [283] and 20 to 60 𝜇m for SIR-Spheres® [284]. 90Y emits high-energy 𝛽− particles (Maximum Energy = 2.28 MeV, Mean Energy = 0.94 MeV) with maximum and mean penetration distances in tissue of 11.3 and 4.1 mm, respectively. This radioisotope is produced by neutron activation of stable 89Y in a nuclear reactor and decays with a half-life of 64.1 h to 90Zr (a stable isotope of zirconium) [285-287]. From a therapeutic standpoint, these properties of 90Y are ideal. For instance, its mean penetration distance in tissue suggests that most of the radiation is deposited a few millimeters away from the MS. Also, its half-life indicates that the radiation does not stay in the body long enough to cause radiotoxic effects. Nonetheless, both commercially available 90Y-MS exhibit undesired characteristics which limit their effectiveness in radioembolization. The problematic features of TheraSphere® and SIR-Spheres® include:   76  Material Limitations  90Y-MS are made of non-biodegradable materials and thus are retained in the patient’s capillary bed indefinitely, preventing re-opening of the embolized capillaries after treatment. They also have a broad size distribution, resulting in deposition of MS in undesired organs (most commonly the lungs) [288]. 90Y-MS are also difficult to administer homogenously due to their high density: 3.3 g mL-1 for TheraSphere® and 1.6 g mL-1 for SIR-Spheres® [289].   Production Limitations  The activity of a radioisotope is the number of atoms that decay in a given time period and is reported in Bq (i.e., becquerel; defined as one disintegration per second) or Ci (i.e., curie; where 1 Ci = 3.7x1010 Bq). At the time of manufacture, 90Y-MS have the following specific activities: 2.5 kBq per MS for TheraSphere® and 40 Bq per MS for SIR-Spheres®. Due to the difference in specific activity, a much higher number of SIR-Spheres® than TheraSphere® is required for a prescribed activity. Furthermore, 90Y-MS are produced only in a few specialized nuclear reactors. Therefore, they often necessitate long transportation times to the hospital during which 90Y decays. This delay results in a dosimetry challenge, as the number of MS required for a prescribed activity increases with time [290].   Imaging Limitations  Although the 𝛽− particles emitted by 90Y have a relatively short range in tissue and high cell killing power, they cannot be directly imaged. The interactions of 𝛽− particles with tissue result in Bremsstrahlung radiation which can potentially be used to create images using gamma cameras [291]. These images, however, have extremely poor resolution and cannot be easily translated into dose estimates due to the lack of quantitative information of 90Y activity distribution. Interestingly, quantitative 90Y Bremsstrahlung might be feasible when advanced data processing methods are applied [292].   Pre-Scan Limitations  Another limitation of TheraSphere® and SIR-Spheres® is the inability of labeling the same MS (glass- and resin-based, respectively) with a diagnostic radioisotope, such as 99mTc (a radioisotope of technetium). A fraction of 90Y-MS may bypass the liver and travel to the lungs, 77  delivering a potentially damaging radiation dose to healthy lung tissue. To avoid treating patients who may suffer this complication, and to rule out deposition of MS in other organs (e.g., the stomach), patients first undergo a hepatic perfusion study using macroaggregated albumin (MAA) labeled with 99mTc (i.e., 99mTc-MAA) as a surrogate of 90Y-MS [293]. This hepatic perfusion study involves simulating the radioembolization procedure with 90Y-MS. Nonetheless, the size distribution of 99mTc-MAA is different than TheraSphere® and SIR-Spheres®, leading to misdiagnosis, inaccurate treatment, and poor patient outcomes [294,295].   In addition to the problems listed above, radioembolization suffers from almost a complete lack of treatment planning and post-90Y administration dosimetry calculations. Even though external beam radiation and brachytherapy plans always include patient-specific radiation dose calculations, the same approach is not routinely utilized in radioembolization. The decision on the activity to be administered for radioembolization is frequently based solely on the weight of the patient [296]. Nevertheless, without proper knowledge of the radiation dose delivered and its distribution, it is very challenging to design an optimal treatment plan for each clinical case and it is impossible to evaluate patient outcomes. Overall, the work described in this chapter aims to develop uniformly-sized MS made of biodegradable and biocompatible polymers in order to overcome the limitations of currently utilized embolic MS and 90Y-MS. To accomplish this, preliminary data is presented on:   1) The development of intrinsically radiopaque rhenium-doped MS (i.e., Re-MS). As extensively discussed in the previous two chapters, rhenium is a strongly radiopaque element and can thus be utilized for X-ray imaging. 2) The development of MS labeled with 188Re (i.e., 188Re-MS). As mentioned in Section 1.2, 188Re emits high energy 𝛽− particles (Maximum Energy = 2.12 MeV, Mean Energy = 0.76 MeV) with maximum and mean penetration distances in tissue of 11.0 and 3.8 mm, respectively. These properties are very similar to those of 90Y, suggesting that 188Re can be used for therapy [297]. However, 188Re also emits 𝛾 photons (Energy = 155 keV with an abundance of 15.6%). These 𝛾 emissions can be imaged by SPECT and can thus be used for dosimetry calculations [1,298]. Transportation problems can be fully avoided because 188Re is produced in a 188W/188Re generator, which can be conveniently placed in hospitals to allow for on-site 188Re labeling of the MS with the exact prescribed activity [6,30].   78  4.1.2 Production of Microspheres by Flow Focusing  Microfluidics revolves around the development of devices to control and manipulate liquids at the microscale [299]. The utilization of microfluidic chips, or lab-on-a-chip, enable the reduction of volumes, times, and costs as well as the overall increase in efficiency and sensitivity [300,301]. Furthermore, they facilitate the miniaturization, integration, and automation of biological and chemical processes and reactions [302].  Generally, two-phase microfluidic flow systems consist of two partially miscible or immiscible fluids that are brought into contact in a microfluidic chip [303]. An example of a two-phase microfluidic flow system which has found applications in diverse pharmaceutical, biomedical, and material sciences applications is the flow focusing method [304].  The flow focusing method allows the production of millions of uniform and size-defined polymer-based spherical droplets with a diameter at the microscale. In a typical flow focusing microchip, a polar continuous phase (CP) flows through two outer microchannels and meets a non-polar dispersed phase (DP) at an inner microchannel, where the DP is squeezed until it breaks into spherical microsized droplets. An example of a flow focusing microchip is shown in Figure 4.1. To facilitate the formation of an emulsion when the two phases are brought into contact, the CP is an aqueous solution of a surfactant (e.g., PVA). The DP is a solution of a polymer, or mixture of polymers, in an organic solvent (e.g., chloroform). The solutions are injected into the microchip using pieces of tubing. The flow rates of both phases are controlled with syringe pumps. The droplets exit the microchip through another piece of tubing. After a single solvent extraction step, the droplets harden and become microspheres [305-307].     Figure 4.1. Flow Focusing Microfluidic Chip. (A) Schematic of the microchip’s flow focusing configuration. The directions of the flows are indicated with red arrows. (B) Image of the microchip fabricated by Bokharaei et al. [305]. This microchip was utilized to produce the MS described in this chapter.  79  The droplets are formed as the result of the Rayleigh instability, which takes place at the interface between two partially miscible or immiscible fluids (like the polar CP and the non-polar DP). This phenomenon occurs primarily as a result of the surface tension of the fluids. Basically, when the two phases are brought into contact, they initially form a thin jet which ultimately breaks into droplets to minimize the surface area. An example of this phenomenon is the formation of droplets when water is dripping from a faucet [308].  The Rayleigh instability, nonetheless, often leads to the production of MS with a bimodal size distribution. Usually, two populations of MS can be distinguished in samples produced by flow focusing: primary and satellite MS. The primary MS represent the main population of MS. The satellite MS constitute only a small fraction of the total number of MS and have a volume much smaller than the primary MS (≪10%). They are derived from satellite droplets (also known as phantom droplets) which are formed as a result of the two main forces which drive the formation of droplets: the viscous force and the interfacial tension force [309,310]. Generally, the size distribution of the MS is important to ensure a homogenous distribution of the MS upon administration and prevent adverse effects associated with the deposition of MS in undesired organs. A sample of MS with a very narrow size distribution is referred to as monosized (or monodispersed). According to the NIST, a sample of MS is monosized if at least 90% of the size distribution lies within 5% of the sample’s mean diameter [311]. The size distribution of a sample of MS can be described in terms of the coefficient of variation (𝐶𝑉), which is defined as the ratio of the standard deviation to the mean diameter and is usually reported as a percentage. Recently, it has been demonstrated that a sample of MS can be considered to be monosized if the 𝐶𝑉 is less than or equal to 3%, which is consistent with the NIST’s monodispersity criterion [312]. The term quasi-monosized (or quasi-monodispersed) has been coined to describe a sample of MS with a 𝐶𝑉 between 3 and 16% [313]. The use of monosized or quasi-monosized MS in biomedical applications is preferred because they have a more predictable biodistribution.   4.2 Materials and Methods  This section is divided into two parts. The first part centers on the development and X-ray imaging of radiopaque Re-MS. The second part focuses on the production of 188Re-MS and the evaluation of their potential application for radioembolization in an HCC-bearing rat model. In 80  both cases, the flow focusing method was used to generate polymer-based MS with a narrow size distribution.   4.2.1 Development and Imaging of Radiopaque Microspheres with Rhenium   Since all various types of hepatic embolotherapy are performed under X-ray imaging guidance and most embolic MS are radiolucent, radiopaque MS are likely to become a new clinical standard. Basically, radiopaque MS have the potential of reducing or avoiding altogether the utilization of I-XCAs in these procedures. Despite the high X-ray attenuation of rhenium, its incorporation into MS has not been reported yet.  The following sections describe a proof of concept study conducted to provide preliminary data regarding the production of Re-MS using a biodegradable and biocompatible polymer doped with rhenium. After characterization, a suspension of these Re-MS was imaged by 𝜇CT.  4.2.1.1 Production and Characterization of the Microspheres  4.2.1.1.1 Production  A tridentate ligand-polymer (or “chelomer”, from chelating polymer) was synthesized followed by rhenium coordination. Based on a procedure reported before by the Hafeli Laboratory, a one pot synthesis of tailored poly(𝐿-lactic acid) (PLA) was conducted, where the resulting polyester chains were capped with a tridentate ligand (2-bis(picolyl)aminoethanol). The chelomer was then coordinated with [Re(H2O)(CO)3]Cl, a rhenium(I) tricarbonyl complex, to make Re-PLA [17].  The newly synthesized rhenium-doped polymer (i.e., Re-PLA) was characterized by IR (where the shift in CO signals is a proof of rhenium coordination), 1H-NMR (which is a proof of PLA backbone and bis(picolyl) head binding), and matrix assisted laser desorption/ionization-time of flight mass spectroscopy, or MALDI-TOF MS (to determine the mass and also confirm the fingerprint of rhenium).  Using this Re-PLA, Re-MS were produced in a flow focusing microchip fabricated and previously reported by Bokharaei et al. (see Figure 4.1B) [305]. A 2% w/v PVA (87-89% hydrolyzed; Sigma Aldrich) solution was prepared as the CP. The weight average molecular weight (𝑀𝑤) of the PVA was 13-23 kDa. The solution was withdrawn into two 10 mL glass syringes assembled with a sterile polyethersulfone membrane filter unit (Pore Size = 0.22 𝜇m, 81  EMD Millipore). The syringes were put in a syringe pump (Fusion 400, Chemyx) placed horizontally and were connected to the CP inlets of the microchip using fluorinated ethylene propylene tubing (ID: 0.05 cm, OD: 0.15 cm; Dolomite). A 5% w/v solution in chloroform of a polymer mixture was prepared as the DP. The polymer mixture was constituted of 97% Re-PLA and 3% MePEG17-b-PCL10, a PEGylated polycaprolactone (PEG-PCL for short). PEG-PCL was added to prevent aggregation of the MS in water, which are mainly constituted of PLA (a hydrophobic polymer). The addition of PEG-PCL at this concentration has been shown to prevent aggregation of similar MS in vivo following administration into mice [314]. After mixing well, this solution was withdrawn into a 1 mL glass syringe, which was then put in another syringe pump (NanoJet, Chemyx) placed vertically. The syringe was finally connected to the DP inlet of the microchip.  The CP and DP flow rates (𝑄𝐶𝑃 and 𝑄𝐷𝑃) were set to 100 and 0.8 𝜇L min-1, respectively. The sample of Re-MS was collected in a 20 mL glass vial with ~4 mL of the 2% w/v PVA solution after the droplet breakup reached steady state conditions. The process was monitored using an inverted microscope (AE31, Motic; Kowloon, China) equipped with a digital camera (Moticam 3, Motic; Kowloon, China). After the allotted collecting time, 260 min, the sample was left overnight in a purifier vertical clean bench (37400-01, Labconco) for chloroform evaporation. To remove PVA traces, the sample was washed with water six times. Specifically, the sample was split into several microcentrifuge tubes with 1 mg of Re-MS each. For each washing step, the Re-MS were suspended in 1 mL of water and mixed by slowly pipetting the suspension up and down five times. The Re-MS were allowed to precipitate for at least 15 min between washing steps. Finally, they were lyophilized (FreeZone®, Lonza) and stored at 4 °C.  Similarly, rhenium-free microspheres (i.e., polymer-based MS without rhenium; referred herein as C-MS, where “C” stands for “control”) were produced under the same experimental conditions as follows: 𝑄𝐶𝑃 = 100 𝜇L min-1 and 𝑄𝐷𝑃 = 0.8 𝜇L min-1. As before, a 2% w/v PVA solution was used as the CP. For the DP, however, a polymer mixture was prepared with 97% of commercially available PLA (𝑀𝑛 = 2,200 Da; Resomer® L104, Boehringer Ingelheim) and 3% of PEG-PCL. The polymer mixture was dissolved at a concentration of 5% w/v in chloroform. The sample of C-MS was collected, recovered, and stored as described above.     82  4.2.1.1.2 Characterization  The size distribution of each sample of MS was determined using a microscopy-based method. The MS were suspended in water at a concentration of 1 mg mL-1. A 20 𝜇L aliquot of each suspension was pipetted onto a glass microscope slide and imaged by brightfield microscopy (AE31, Motic; Kowloon, China). The diameters of more than 500 MS in several images were measured using ImageJ.  The external morphology of the MS was inspected by SEM (SU3500, Hitachi; Chiyoda, Japan). Using the suspensions prepared before, 20 𝜇L were pipetted onto a specimen stub covered with an electrically conductive carbon-based adhesive disc. The specimen stubs were left in a purifier vertical clean bench (37400-01, Labconco) until the water evaporated from the suspensions (>8 h), leaving behind just the samples of MS. Once fully dried, the samples were sputter-coated with an 8 nm layer of iridium using a modular high vacuum coating system (EM MED020, Leica; Wetzlar, Germany) under reduced pressure (<5 Pa). The images were acquired at 15 kV with a magnification of 1,000x.   4.2.1.2 X-Ray Imaging  The samples of Re-MS and C-MS were imaged by 𝜇CT using a preclinical 𝜇CT scanner (eXplore CT120, TriFoil Imaging; Chatsworth, USA).  The MS were suspended in water at a concentration of 11.5 mg mL-1. This is the maximum concentration at which the MS remain in suspension for ~15 min, which is the estimated duration of the scan. Then, 200 𝜇L of the suspensions of Re-MS and C-MS were pipetted into two microcentrifuge tubes, which were placed in the phantom described in Section 2.2.2.1. The following samples were also placed in the same phantom: 1) 200 𝜇L of an ammonium perrhenate solution with a concentration of rhenium of 200 mM, 2) 200 𝜇L of water, and 3) air (i.e., an empty microcentrifuge tube).  The scan was performed at 50 kVp (100 mA and 100 ms). As mentioned in Section 2.2.2.1, the eXplore CT120 (TriFoil Imaging; Chatsworth, USA) has an inherent filtration of 1 mm of aluminum and 0.15 mm of beryllium. A 0.2 mm thick copper attenuator was utilized as additional filtration. The FOV had a rectangular shape with length and width of 8.5 x 5.5 cm, respectively. The acquisition consisted of 1,440 projections of the phantom. The scanner was operated in the step-and-shoot mode with a 3 s step delay. Using MicroView (Parallax 83  Innovations; Ilderton, Canada), the data was reconstructed as a 3D image with a matrix size of 1455 x 1455 and a spatial resolution and a slice thickness of 100 𝜇m.  Due to the small volume of the formulations in the microcentrifuge tubes, the majority of the axial images in the 3D image showed a significant amount of air on top of the samples. A representative axial image was thus chosen in order to calculate the mean pixel intensity of each sample. Using Image-Pro Plus 7 (Media Cybernetics; Rockville, USA), a line-of-interest (LOI) with 20 pixels in length was drawn for each sample and the distribution of grey values of all the pixels in the LOI was determined. Additionally, a square-shaped ROI with sides of 25 pixels was drawn for each sample and a 3D plot of the distribution of grey values was created (see Figure 4.11A in Section 4.3.1.2). This analysis was performed to corroborate the uniformity of the contrast of the samples in the axial image.   4.2.2 Development and Imaging of Radioactive Microspheres with Rhenium-188  Following the characterization of Re-MS, the suitability of the flow focusing method to produce uniformly-sized MS made of mixture of biodegradable and biocompatible polymers became evident. The size of the MS can be easily adjusted by varying the CP and DP flow rates. For example, larger MS can be generated at lower 𝑄𝐶𝑃 and/or higher 𝑄𝐷𝑃 [305]. Using the flow focusing method, ~40 𝜇m MS with a narrow size distribution were then produced. To turn these MS radioactive, they were labeled with 188Re (see Section 4.2.2.2). The feasibility of using these newly developed 188Re-MS for radioembolization in an HCC-bearing rat model was evaluated in a small-scale pilot study. The objective of this study was to garner initial data of the biodistribution of 188Re-MS in Sprague-Dawley rats undergoing a radioembolization procedure. A total of four rats were randomized into two cohorts: Cohort A (HCC/+, or HCC-bearing rats) and Cohort B (HCC/-, or healthy rats).  HCC was induced in the rats allocated to Cohort A by injecting them directly into the liver with a suspension of N1-S1 cells (ATCC® CRL-1604TM; Manassas, USA). These commercially available cells were established by administration of 4-dimethylaminoazobenze, a carcinogen, into a male Sprague-Dawley rat [315]. Upon inoculation in the liver, these cells proliferate and form a highly chemoresistant tumor known as Novikoff HCC [30], which mimics the resistance often seen in human HCC (see Section 4.2.2.3) [316]. The four rats were subjected to a hepatic intra-arterial catheterization for administration of 188Re-MS 2 weeks post-inoculation of the cells. 84  The biodistribution of 188Re-MS was appraised in images acquired by SPECT 1, 24, 48, and 72 h post-radioembolization (see Section 4.2.2.4). This pilot study is mainly comprised of three stages: 1) Production and Characterization of 188Re-MS, 2) Development of an HCC-Bearing Animal Model, and 3) Radioembolization and Quantitative SPECT Imaging. These stages are described in detail in the next few sections.   4.2.2.1 Animals  Four ~250 g male Sprague-Dawley rats were purchased from Charles River Laboratories (Wilmington, USA) and housed at the Modified Barrier Facility (MBF) and the Centre for Comparative Medicine (CCM) at UBC in Vancouver, Canada. The rats were provided with food and water ad libitum. UBC’s Animal Care Committee approved protocol A15-0244. All animal procedures were conducted in accordance with the guidelines issued by the Canadian Council on Animal Care (CCAC).  4.2.2.2 Production, Radiolabeling, and Characterization of the Microspheres  4.2.2.2.1 Production  Using the microchip depicted in Figure 4.1B, MS containing a chelomer for [188Re(CO)3]+ binding were produced by flow focusing. The experimental conditions were chosen to generate ~40 𝜇m MS with a narrow size distribution. The CP was a 2% w/v PVA (𝑀𝑤 = 13-23 kDa, 87-89% hydrolyzed; Sigma Aldrich) solution. The DP was a 20% w/v solution in chloroform of a polymer mixture. The polymer mixture was constituted of 87% commercially available PLA (𝑀𝑛 = 2,200 Da; Resomer® L104, Boehringer Ingelheim), 10% of the custom-synthesized chelomer described in Section 4.2.1.1.1: PLA capped with 2-bis(picolyl)aminoethanol (a tridentate ligand), and 3% PEG-PCL (added to prevent aggregation of the MS in water [314]).  The sample was collected for 420 min at a 𝑄𝐶𝑃 of 30 𝜇L min-1 and a 𝑄𝐷𝑃 of 1 𝜇L min-1. To remove PVA traces, the MS were washed with water six times as described before. They were then lyophilized (FreeZone®, Lonza) and stored at 4 °C until they were needed.        85  4.2.2.2.2 Radiolabeling  A 188W/188Re generator (ITM Isotopen Technologien München AG; Garching, Germany) was utilized as the source of [188ReO4]-, which was eluted as Na(188ReO4) using 0.9% w/v sodium chloride (NaCl, or simply saline). To label the MS with 188Re, a customized 188Re labeling kit was prepared to reduce [188ReO4]- to [188Re(H2O)3(CO)3]+ under vacuum employing potassium boronocarbonate as both a reducing agent and an in situ source of CO. [188Re(H2O)3(CO)3]+ was added to a 20 mg mL-1 suspension of MS in water. The activity of [188Re(H2O)3(CO)3]+ added was 67.7 MBq per mg of MS. The mixture was incubated at 70 °C for 30 min in a Thermomixer R (Eppendorf; Mississauga, Canada) shaking at 500 rpm. The 188Re-MS were then allowed to precipitate (≥10 min), the supernatant was removed, and the 188Re-MS were suspended in 50 𝜇L of saline.  The labeling efficiency was calculated by dividing the total activity minus the activity of the supernatant over the total activity.   4.2.2.2.3 Characterization  The size distribution of the MS before [188Re(CO)3]+ binding was determined using a microscopy-based method. The diameters of more than 500 MS in several images acquired through brightfield microscopy (AE31, Motic; Kowloon, China) were measured using ImageJ. The external morphology of the MS before and after [188Re(CO)3]+ binding was inspected by SEM (SU3500, Hitachi; Chiyoda, Japan) at 15 kV with a magnification of 1,000x. For both of these techniques, the preparation of the samples was performed as per described in Section 4.2.1.1.2.  4.2.2.2.4 Development of a Parenteral Formulation  For administration into the rats during the radioembolization procedure, a parenteral formulation was developed by suspending the 188Re-MS in a sterile solution made of 1.75% w/v hydroxyethyl cellulose (HEC; 𝑀𝑤 = 24-27 kDa; Polysciences Inc.), 5% w/v dextrose (Baxter Corporation), and 0.01% w/v Chicago Sky Blue 6B (Sigma Aldrich) (hereafter abbreviated as HEC/Dex/CSB). HEC was used as a viscosity enhancer, whereas dextrose was used as a tonicity agent. Chicago Sky Blue 6B, a dye, was simply added to enhance the visibility of the 188Re-MS during administration. The solution was autoclaved for sterilization.  86  The viscosity and the osmolality of the formulation were measured using a rotational rheometer (MCR 301, Anton Paar; Graz, Austria) and a vapor pressure osmometer (VAPRO®, EliTech Group; Puteaux, France), respectively.   4.2.2.3 Development of a Hepatocellular Carcinoma-Bearing Animal Model   Before administration of 188Re-MS, tumors were grown by injecting a suspension of N1-S1 cells directly into the liver of two rats using ultrasound guidance. From the four rats utilized in this pilot study, two rats (identified as R01 and R02) received an injection of cells and developed a tumor, which was corroborated in a post-mortem examination. This is Cohort A, or HCC/+. The remaining two rats (identified as R03 and R04) were healthy at the time of radioembolization. This is Cohort B, or HCC/-.    4.2.2.3.1 Culture and Preparation of Doses of N1-S1 Cells  The cells were cultured in suspension at 37 °C and 5% CO2 in Iscove’s Modified Dulbecco Medium (IMDM; ThermoFisher Scientific) supplemented with 10% w/v FBS (ThermoFisher Scientific) and 1% w/v Pen-Strep (10,000 U mL-1, ThermoFisher Scientific). The initial culture was started at 4x104 viable cells per mL and all the subcultures were maintained between 1x105 and 5x105 viable cells per mL.  A Trypan Blue exclusion test was performed to distinguish between viable and dead cells. If cells take up Trypan Blue, they are considered dead (their membrane is compromised). An aliquot of a single cell suspension (also known as a homogenous cell suspension) was diluted 1:1 with a 0.4% w/v Trypan Blue solution (ThermoFisher Scientific). The unstained cells (i.e., viable cells) were counted using a hemocytomer grid. Each dose was prepared with 1x106 cells, which has been reported to form a ~0.5 to 1 g tumor 2 weeks post-inoculation of the cells [30,317].  Before inoculation, the cells were suspended in 50 𝜇L of CorningTM MatrigelTM Membrane Matrix (MatrigelTM for short; Fischer Scientific). MatrigelTM is a mixture of proteins which favors cell attachment and proliferation. Interestingly, MatrigelTM irreversibly polymerizes at room temperature, producing a stable and effective matrix for the attachment and differentiation of cells [318]. To avoid polymerization of the suspension of cells before injection, the syringes were kept on ice. After injection, however, MatrigelTM polymerizes immediately. This has been reported to form single tumors in mice and rats at the site of injection, markedly preventing the 87  migration of cells and thus the formation of secondary tumors [319-321]. Each dose was prepared in a 1 mL syringe connected to a 30G 1’’ needle.   4.2.2.3.2 Ultrasound-Guided Inoculation of N1-S1 Cells into the Liver  The inoculation of cells was performed via an ultrasound-guided percutaneous injection using the Vevo® 2100 Digital Imaging Platform (Visual Sonics; Toronto, Canada) (Figure 4.2). A 40 MHz ultra-high frequency transducer, also from Visual Sonics, was utilized to visualize the liver and inject the cell suspensions directly below the capsule of the left lobe of the liver.     Figure 4.2. Ultrasound Machine. The instrument can operate in B-mode (from brightness mode, or 2D imaging) and M-mode (from motion mode, or time-resolved imaging) with a resolution of 30 𝜇m and frame rates up to 740 fps. Furthermore, quantification of blood flow is feasible on PW-mode (from pulsed wave Doppler mode). Model and Manufacturer: Vevo® 2100 Digital Imaging Platform, Visual Sonics (Toronto, Canada). Location: Faculty of Pharmaceutical Sciences at UBC in Vancouver, Canada.       The rats were anesthetized with isoflurane (Fresenius Kabi) and mounted in a supine position on the machine’s heated operating table with continuous monitoring of vital signs (Figure 4.3). As preemptive analgesia, meloxicam (1 mg kg-1) was administered via a subcutaneous injection (SQ). The abdomen was shaved, disinfected with 0.5% w/v chlorhexidine solution (Stanhexidine®, Omega), and fully covered with a thin layer of ultrasound gel (Aquasonic®, Parker Laboratories Inc.). The liver was visualized on the screen. The syringe filled with the cell suspension was brought to the skin just below the rib cage at an angle of 20 to 40° with the bevel up. After visualization on the screen, the needle was passed through the skin and the abdominal wall muscles and it was then directed to the liver. The cells were slowly 88  injected into the left lobe. After a 3 s pause to prevent leakage of the cells, the needle was rapidly withdrawn. The rats were immediately recovered.     Figure 4.3. Setup for Ultrasound-Guided Inoculation of N1-S1 Cells into the Liver. The rats were anesthetized with isoflurane, mounted on a heating operating table, and shaved. The injections were done at an angle of 20 to 40° using a 30G 1’’ needle. The process was monitored live on the screen of the instrument. The image shows (a) the injection system, (b) the transducer, and (c) the operating table.     4.2.2.4 Radioembolization and Quantitative SPECT Imaging   To allow for tumor growth in the two rats that underwent the procedure described in the previous section (Cohort A, or HCC/+), the hepatic intra-arterial catheterization for 188Re-MS administration was performed 2 weeks post-inoculation of the N1-S1 cells. To evaluate the differences in the biodistribution of the 188Re-MS, this surgery was also performed in the two healthy rats (Cohort B, or HCC/-). A total of 2 to 3 mg of 188Re-MS suspended in 100 𝜇L of HEC/Dex/CSB solution were administered into the rats. Table 4.1 summarizes the number of 188Re-MS and the activity injected in each rat.          89  Table 4.1. Dose of 188Re-Labeled Microspheres per Rat. Cohort A: HCC/+. Cohort B: HCC/-. The rats in Cohort A received an injection of 1x106 N1-S1 cells in 50 𝜇L of MatrigelTM directly into the liver using ultrasound guidance. All the rats received an injection of 188Re-MS via a hepatic intra-arterial catheterization 2 weeks post-inoculation of the cells in the rats allocated to Cohort A. R01, R02, and R04 received an injection of 2 mg of 188Re-MS. R03 received an injection of 3 mg of 188Re-MS.  Cohort  ID Animal  ID Number of  188Re-MS  Activity  (MBq) Specific Activity (Bq per MS) Cohort A R01 47,700 44.4 945 R02 47,700 12.6 238 Cohort B R03 71,600 55.5 782 R04 47,700 11.5 245   The number of 188Re-MS injected was estimated by dividing the amount of 188Re-MS injected by the weight per 188Re-MS (i.e., 4.19x10-5 mg). The weight per 188Re-MS was calculated by multiplying the volume per 188Re-MS (calculated considering a diameter of 40 𝜇m) times the density of PLA (reported to be 1.2 g mL-1 [322]). The activity injected was determined by calculating the difference in the activity measurements of the catheters, needles, and syringes before and after 188Re-MS administration. The activity measurements were performed with a dose calibrator (AtomlabTM 500, Biodex Medical Systems Inc.; Shirley, US).      4.2.2.4.1 Administration Device  For each rat, a vascular catheter made of polyethylene (PE-50, ID = 0.05 cm and OD = 0.10 cm; IntramedicTM, Becton Dickinson) was customized to administer 188Re-MS with minimal losses in the catheters, needles, and syringes. To do this, one end of a 20 cm long PE-50 catheter (referred to as “Catheter I”) was stretched by applying heat so that its final size was approximately equal to that of a PE-10 catheter (ID: 0.02 cm, OD: 0.05 cm). This step was done with the help of a heat gun, which was placed approximately 30 cm away from the catheter. To stretch the catheter, it was twisted and pulled simultaneously after turning on the heat gun. The tip of the catheter was cut at a 45° angle with a razor blade to facilitate insertion into the rat. The stretched section was approximately 5 cm long. The unaltered end of Catheter I was coupled to a 1 mL syringe filled with 500 𝜇L of saline and connected to a 23G 1’’ blunt needle (referred to as “Syringe A”). A 100 cm long PE-50 catheter (referred to as “Catheter II”) was connected to a 1 mL syringe filled with 200 𝜇L of saline (referred to as “Syringe B”) and was placed inside a 3 mm thick box made of PMMA to shield the 𝛽− emissions of 188Re. This system was then put inside a lead container which shielded the 𝛾 emissions of 188Re from one side (Figure 4.4). A ~1 90  cm thick sheet of lead was placed on the other side of the container to shield the 𝛾 emissions from the other side.     Figure 4.4. Device for Administration of 188Re-Labeled Microspheres. The PMMA box holds Catheter II, which is 100 cm long and is loaded with the suspension of 188Re-MS in HEC/Dex/CSB. This box shields the 𝛽− emissions of 188Re. This system is placed in a lead container which shields the 𝛾 emissions of 188Re from one side. To fully shield the 𝛾 emissions, a ~1 cm thick sheet of lead was placed on the other side of the container. Syringe B is filled with 200 𝜇L of saline utilized to flush Catheter II after being assembled with Catheter I, used to catheterize the rat.      The 188Re-MS were suspended in 100 𝜇L of the HEC/Dex/CSB solution described in Section 4.2.2.2.4. This suspension was withdrawn into Catheter II, followed by 50 𝜇L of saline. The catheter is long enough to contain this volume without drawing the 188Re-MS into the syringe, where they could get stuck (leading to losses during administration). A polished 1 cm long piece of a 23G blunt needle was then inserted into the free end of Catheter II (Figure 4.5). This assemble is referred herein as the 23G/Catheter II/Syringe B system.     Figure 4.5. Set of Catheters for Administration of 188Re-Labeled Microspheres. Catheter II is used as a reservoir for the suspension of 188Re-MS in HEC/Dex/CSB. Catheter II is 100 cm, which is long enough to hold a volume of 100 𝜇L. Using this set of catheters, the 188Re-MS never reach the syringe, where losses may occur. The image was adapted from [30].       91  4.2.2.4.2 Hepatic Intra-Arterial Catheterization  The suspension of 188Re-MS in HEC/Dex/CSB was administered directly into the proper hepatic artery (PHA), which provides the tumor with most of its blood supply. The PHA arises from the common hepatic artery (CHA) and, together with the hepatic portal vein and the common bile duct, forms part of the portal triad. The portal triad is a distinctive group of blood vessels in the liver. Basically, radioembolization with 188Re-MS was accomplished by performing a laparotomy followed by a hepatic intra-arterial catheterization.  The rats were anesthetized with isoflurane and mounted in a supine position on a heated operating table with continuous monitoring of vital signs. The abdomen was shaved and disinfected with 0.5% w/v chlorhexidine solution (Stanhexidine®, Omega) and 70% w/v isopropyl alcohol. As a local anesthetic, 8 mg kg-1 of a 0.25% w/v bupivacaine solution was injected on the incision site, between the right and the left upper quadrants of the abdomen. A laparotomy was then performed and the left lateral lobe of the liver was externalized and retracted using a piece of gauze soaked with saline and held back by needle holders. Cotton tipped applicators were used to gently dissect through the mesentery to expose the portal triad after pushing the first loop of the duodenum caudally. A clean cotton-tipped applicator was utilized to clear the portal triad of excess connective tissue and expose the following blood vessels: PHA, CHA, and gastroduodenal artery (GDA). To prevent bleeding in the event of puncture, the CHA was temporarily ligated with a removable micro bulldog clamp. A small curved hemostat was used to isolate the GDA. To control retrograde bleeding, the GDA was ligated with a 4-0 suture close to the branch point with the posterior pancreatinoduodenal arcade. A 2-0 suture beneath the GDA was utilized to apply tension on the blood vessel during the catheterization (Figure 4.6).   92    Figure 4.6. Schematic of the Hepatic Intra-Arterial Catheterization. The cartoon depicts the most important blood vessels involved in the surgery performed for radioembolization of the liver with 188Re-MS. Access to the PHA was gained through the GDA. During the surgery, blood flow in the CHA was temporarily stopped in the CHA with a micro bulldog clamp and in the GDA with a 4-0 suture. These two ligation sites were repaired after administration of the 188Re-MS to ensure appropriate blood flow to the liver.   A surgical microscope (M525 F20, Leica; Wetzlar, Germany) was used to insert the stretched end of Catheter I into the GDA proximal to two ligation sites, but as distally to the branch point of the GDA from the CHA as possible. While maintaining tension on the 2-0 suture under the GDA, the catheter was carefully advanced into the CHA, placing the tip in the common branch with the PHA. The catheter was then fixed in place with a 2-0 suture. The knot was placed proximal to the insertion site to prevent bleeding and hold the catheter tightly within the wall of the PHA without occluding the lumen. When Catheter I was secured at the injection site, Syringe A was exchanged with the 23G/Catheter II/Syringe B system.  The previously externalized liver lobe was returned to the abdominal cavity. To restore hemodynamics, 100 𝜇L of heparin lock flush and 500 𝜇L of saline were infused. After removal of Catheter I, the two ligation sites were repaired with 2-0 sutures and the presence of appropriate blood flow to the liver was confirmed. The body wall and skin were closed with 5-0 sutures with simple and continuous subcutaneous patterns, respectively.  All sutures were made with Coated VICRYL® (Ethicon), a biodegradable material composed of a copolymer of poly(glycolic acid) and PLA. The rats received a split dose of buprenorphine (i.e., half dose pre-op and half dose post-op; Total Dose = 0.05 mg kg-1 SQ). 93  Upon recovery, they all continued on buprenorphine (0.01-0.05 mg kg-1 SQ). The dose was determined on an individual basis depending upon the degree of pain signs and discomfort. Additionally, meloxicam (1 mg kg-1 SQ) was administered if deemed necessary following a clinical assessment.    4.2.2.4.3 Quantitative SPECT Imaging and Dosimetry  Still under general anesthesia from the surgery described above, the rats were imaged using the VECTor Imaging Platform (MILabs; Utrecht, Netherlands), a preclinical SPECT/PET/CT system (Figure 4.7). Taking into account the time required to close the sutures and stabilize the vital signs of the rats, the scans were performed approximately 1 h post-radioembolization. To assess the biodistribution of 188Re-MS over time, additional SPECT/CT scans were performed 24, 48, and 72 h post-radioembolization.     Figure 4.7. VECTor Imaging Platform. This system is capable of imaging both SPECT and PET radiotracers with a submillimeter spatial resolution. Furthermore, the CT module provides anatomical information of the animals and is utilized to create transmission maps for use in attenuation correction [4]. Model and Manufacturer: VECTor Imaging Platform, MILabs (Utrecht, Netherlands). Location: CCM in Vancouver, Canada.      The rats were placed on a heating pad during the scans with continuous monitoring of vital signs. The SPECT scans were performed with the following rat size multi-pinhole collimator: Ultra-High Resolution Collimator, simply referred to as UHRC. This collimator has these characteristics: Bore Diameter = 98 mm, Wall Thickness = 15 mm, Pinhole Diameter = 1 mm, Number of Pinholes = 75 [4]. The scan time was increased for the later time point scans to maximize the counts and minimize the noise, but taking into account the time that the rats could remain safely under anesthesia. For example, the scan time was 30 min for the scans conducted 1 94  h post-radioembolization, 60 min for the scans conducted 24 and 48 h post-radioembolization, and 90 min for the scans conducted 72 h post-radioembolization. After the SPECT scans, whole-body CT scans were conducted at 45 kVp (615 𝜇A).  The CT images were utilized to obtain anatomical information of the rats and create a transmission map which was used for attenuation correction. The SPECT images were corrected for attenuation and scatter, which was specifically accomplished with the triple-energy window method [323]. Using AMIDE’s Medical Imaging Data Examiner (or AMIDE for short, an open-source image processing software), the tumor-to-healthy liver tissue (𝑇/𝐻) was determined by calculating the ratio of the mean voxel intensity in the tumor to the mean voxel intensity in the healthy liver tissue. The mean voxel intensities in the target regions were determined by drawing spherical VOIs in the corresponding tumor and healthy liver tissue volumes on SPECT/CT images and calculating their mean voxel intensities. The number of VOIs was selected to accurately represent the complete volume of the target regions. For example, up to three VOIs with 4 pixels in diameter were required for the tumor, but eight VOIs with 10 pixels in diameter were required for the healthy liver tissue. The radiation doses to the tumor and the healthy liver tissue were estimated for each rat with Monte Carlo using the dose kernel convolution method. For this, a 188Re point source dose kernel was simulated with Monte-Carlo. The point-dose kernel was convolved with a simple model of the rat tumor/liver activity map. This model included information about the following parameters: 𝑇/𝐻, activity injected, and sizes of the tumor and the liver. A simplified model of the rat tumor/liver activity map was convolved with the point-dose kernel. In this simplified model, the rat liver geometry was modeled as a uniform sphere made of liver tissue (the density was assumed to be 1.06 g mL-1). Similarly, the liver tumor was modelled as a second sphere, placed off-center inside the liver. The sizes of the entire liver and the tumor models were equal to those measured experimentally. The activity map was created by distributing uniformly the measured injected activity in normal liver and tumor volumes based on the measured 𝑇/𝐻 values.  4.3 Results and Discussion  The following sections summarize the main experimental findings derived from the production and imaging of uniformly-sized polymer-based Re-MS and 188Re-MS. 95  4.3.1 Development and Imaging of Radiopaque Microspheres with Rhenium  4.3.1.1 Production and Characterization of the Microspheres  The newly produced Re-MS were found to have a diameter of 14.6 ± 2.8 𝜇m. The histogram depicted in Figure 4.8A shows that the sample is composed of Re-MS with a diameter between 5 and 20 𝜇m, but in fact more than 90% of the Re-MS have a diameter larger than 10 𝜇m. Based on the NIST’s monodispersity criterion [311-313], with a 𝐶𝑉 of 18.8%, the Re-MS are almost quasi-monosized. Figure 4.8B illustrates a representative image acquired by SEM, where it is evident that the Re-MS have a spherical shape and a smooth surface. While produced under the same experimental conditions, the C-MS (i.e., polymer-based MS without rhenium) were found to be slightly smaller than the Re-MS (C-MS: 10.6 ± 1.3 𝜇m, 𝐶𝑉 = 12.3%; Re-MS: 14.6 ± 2.8 𝜇m, 𝐶𝑉 = 18.8%). The decrease in size is attributed to the use of different polymers to prepare the samples. The C-MS were constituted predominately of commercially available PLA (𝑀𝑛 = 2,200 Da; Resomer® L104, Boehringer Ingelheim), whereas the Re-MS were constituted predominately of Re-PLA (𝑀𝑛 = 2,524 Da, determined by MALDI TOF-MS after rhenium coordination).     Figure 4.8. Size Distribution and Morphology of the Rhenium-Doped Microspheres. (A) The diameters of at least 500 Re-MS were measured in images acquired by brightfield microscopy (AE31, Motic; Kowloon, China). The Re-MS were found to be 14.6 ± 2.8 𝜇m in diameter (𝐶𝑉 = 18.8%). (B) Image of the Re-MS acquired by SEM.    As it was vastly discussed in Chapter 2, the X-ray attenuation of a material is dependent on the concentration of the radiopaque element. In Section 2.3.1.1, this was experimentally demonstrated by imaging by 𝜇CT rhenium- and iodine-based formulations with a concentration of rhenium and iodine between 50 and 200 mM. A positive correlation was found between the 96  𝐶𝑁𝑅s of the samples and the concentration of rhenium and iodine (see Figure 2.14). The concentration of rhenium in the suspension of Re-MS (reported in mM) depends on the following variables: 1) the mole percent rhenium content in the custom-synthesized Re-PLA, 2) the mole percent rhenium content in the Re-MS, and 3) the concentration of Re-MS in the suspension (reported in mg mL-1). First, the mole percent rhenium content in Re-PLA was calculated by dividing the atomic number of rhenium (i.e., 186.2 Da) by the molecular weight of Re-PLA (i.e., 2,524 Da). The quotient was multiplied by 100 to be expressed as a percentage. The structure of this custom-synthesized rhenium-doped polymer is shown in Figure 4.9. Thus, the mole percent rhenium content in Re-PLA is 7.4%. Second, the mole percent rhenium content in the Re-MS was calculated by multiplying the mole percent rhenium content in Re-PLA (i.e., 7.4%) by the concentration of Re-PLA in Re-MS (i.e., 97%, see Section 4.2.1.1.1). Thus, the mole percent rhenium content in Re-MS is 7.2%. Last, the concentration of Re-MS in the suspension was 11.5 mg mL-1 (see Section 4.2.1.2) Considering these last two variables, the concentration of rhenium in the suspension was calculated as follows: (11.5 / 1000 g of Re-MS) (7.2 / 100) / 186.2 g mol-1 / (1 / 1000 L). The ensuing value was multiplied by 1,000 to express the result in mM. Hence, the concentration of rhenium in the suspension is 4.4 mM.      Figure 4.9. Structure of Rhenium-Functionalized Polymer. The image shows the structure of a PLA chain capped with a bis(picolyl) head after coordination with [Re(CO)3]+. This structure was first reported in [17].   4.3.1.2 X-Ray Imaging  Figure 4.10 shows a representative axial image of the samples in the phantom. The suspension of Re-MS has a slightly higher attenuation than the suspension of C-MS (Re-MS: 51 HU; C-MS: 8 HU). As expected, the solution of ammonium perrhenate displays a significant increase in contrast, with an attenuation of 1,742 HU. The low attenuation of the suspension of Re-MS compared to the solution of ammonium perrhenate is attributed to the low concentration of rhenium in the suspension of Re-MS. For instance, the concentration of rhenium in the 97  solution of ammonium perrhenate is ~45 times higher than in the suspension of Re-MS (200 and 4.4 mM, respectively). The suspension of Re-MS was prepared with the maximum amount of Re-MS which can remain in suspension for the entire duration of the scan (~15 min). If Re-MS start to precipitate during the scan, then the contrast will not be uniform in the images. Thus, increasing the concentration of Re-MS in the suspension was simply not feasible.     Figure 4.10. Micro-Computed Tomography Imaging. The image shows a representative axial image acquired at 120 kVp (10 mAs) by 𝜇CT (eXplore CT120, TriFoil Imaging; Chatsworth, USA). The following samples are shown: (a) air, (b) C-MS suspension, (c) Re-MS suspension ([Re] = 4.4 mM), (d) ammonium perrhenate solution ([Re] = 200 mM), and (e) water. Attenuation: (a) -1,000 HU, (b) 8 HU, (c) 51 HU, (d) 1,742 HU, and (e) 0 HU.   As described in Section 4.2.1.2, Image-Pro Plus 7 (Media Cybernetics; Rockville, USA) was used to calculate the mean pixel intensities of the samples (using LOIs) and create 3D plots of the distribution of grey values in the samples (using ROIs) (Figure 4.11A). Figure 4.11B shows that there is no significant variation in the grey values across the LOIs. This provides evidence that both Re-MS and C-MS successfully remained in suspension during the scan. Also, 3D plots of the distribution of grey values in square-shaped ROIs with sides of 25 pixels are illustrated in Figures 4.11C-G. For the Re-MS suspension, the mean pixel intensity was found to be 43.2 ± 0.7, whereas for the ammonium perrhenate solution, it was found to be 83.6 ± 0.4. The net mean pixel intensity of the Re-MS suspension, calculated by subtracting the mean pixel intensity of the background (i.e., 42.2 ± 0.7), is 1.0. By extrapolation, a suspension of Re-MS with a concentration of rhenium of 200 mM will thus have a net mean pixel intensity of 45.8. The net mean pixel intensity of the ammonium perrhenate formulation is 41.4, which agrees within 10% of the theoretical and the experimental values. At the same time, it suggests that, if the concentration of rhenium is increased, the suspension of Re-MS will exhibit a significant increase in contrast.   98    Figure 4.11. Pixel Intensity Analysis. (A) Cartoon of the quantitative LOI- and ROI-based image analysis. The circles represent the samples. The black lines inside the circles represent the LOIs, which were 20 pixels in length (or 2 mm long taking into account that the resolution of the image was 100 𝜇m). The black squares surrounding the circles represent the ROIs used to create the 3D plots, with sides of 25 pixels. (B) The graph shows that the contrast of the samples is uniform across the LOIs. (C-G) 3D plots of the distribution of grey values in the samples. Samples: (C) air, (D) C-MS suspension, (E) Re-MS suspension ([Re] = 4.4 mM), (F) ammonium perrhenate solution ([Re] = 200 mM), and (G) water.  4.3.2 Development and Imaging of Radioactive Microspheres with Rhenium-188  4.3.2.1 Production, Radiolabeling, and Characterization of the Microspheres  The MS have a diameter of 41.8 ± 6.0 𝜇m. The sample, however, is constituted of two populations of MS: primary MS (42.7 ± 1.4 𝜇m, 𝐶𝑉 = 3.3%) and satellite MS (15.5 ± 3.3 𝜇m, 𝐶𝑉 = 21.5%) (Figure 4.12). One of the problems with current radioembolization agents is that the MS have a broad size distribution: 20 to 30 𝜇m for TheraSphere® [283] and 20 to 60 𝜇m for SIR-Spheres® [284]. Due to their broad size distribution, 90Y-MS are usually not distributed homogenously within the tumor. This in turn results in a non-homogenous radiation therapy [324]. Even with the presence of satellite MS, the sample exhibits a fairly narrow size distribution with a 𝐶𝑉 of 14.5%. Thus, the MS are quasi-monosized.  99   Figure 4.12. Size Distribution of the Microspheres. The diameters of at least 500 MS were measured in images acquired by brightfield microscopy (AE31, Motic; Kowloon, China). The MS exhibit a bimodal size distribution. The MS were found to be 41.8 ± 6.0 𝜇m in diameter (𝐶𝑉 = 14.4%).    The satellite MS can be separated by sieving the sample. Owing to the fact that the relative frequency of the satellite MS is negligible compared to the parent MS (<4%), the sample was utilized in subsequent steps without sieving it. The preparation of MS by flow focusing thus successfully eliminates unnecessary recovery and purification steps, which are typically required when other methods of preparation of MS are employed (e.g., precipitation polymerization, dispersion polymerization, and emulsion polymerization) [312].   As shown in Figure 4.13, the MS are highly spherical and they preserve their size and shape after labeling with 188Re. A schematic of the 188Re labeling step is shown in Figure 4.14. The MS were functionalized with [188Re(CO)3]+ with a labeling efficiency of 50 to 80% across different batches. Free [188Re(CO)3]+ (i.e., not bound to MS) was fully removed by allowing the MS to precipitate, removing the supernatant, and washing the MS with water several times. Thus, 100% of the activity injected into the animals corresponded to MS labeled with [188Re(CO)3]+. Furthermore, a thin layer chromatography (TLC) showed more than 95% [188Re(CO)3]+ binding stability over 24 h.   100    Figure 4.13. Morphology Inspection of Microspheres Before and After 188Re Labeling. Images acquired by SEM are shown for the MS (A) before and (B) after 188Re labeling. For (B), the images were taken after 188Re decay. No morphological changes were observed in the MS after 188Re labeling.    Figure 4.14. Schematic of the Surface of the 188Re-Labeled Microspheres. The image shows the surface of a single MS with the chelomer before and after labeling with [188Re(CO)3]+.    Several clinical studies have shown that TheraSphere® and SIR-Spheres® can escape the liver and travel to the lungs, triggering deleterious radiotoxic effects. Moreover, both of these radioembolization agents are retained in the patient’s capillary bed indefinitely as a consequence of their non-biodegradable nature, thus preventing reopening of the clogged capillaries after radioembolization [325-327]. The newly developed 188Re-MS constitute the first effort to produce fully biodegradable and uniformly-sized MS bound to 188Re for radioembolization. Due to the slightly larger size of 188Re-MS compared to 90Y-MS, they are also expected to have an improved embolic effect. Ultimately, this has the potential of avoiding adverse effects experienced by patients with a high degree of shunting to the lungs (e.g., restrictive ventilatory dysfunction and radiation pneumonitis [327]).  101  As a consequence of the high density of TheraSphere® and SIR-Spheres® (3.3 and 1.6 g mL-1, respectively [289]), 90Y-MS are difficult to administer homogenously. This has been associated with: 1) settling of the MS before and during administration in the lumen of the catheters and 2) suboptimal distribution of the MS within the tumor-feeding blood vessels [328,329]. 188Re-MS, however, are mainly made of PLA, which has a density of 1.2 g mL-1 [322]. An ideal radioembolization agent should have a density close to blood [290], which is nearly equal to water [330]. Due to the significantly lower density of 188Re-MS compared with both of the two currently utilized 90Y-MS, 188Re-MS are expected to exhibit improved flow dynamics. Based on clinical experience, this could strongly enhance localization of the MS within the tumors, which will in turn improve patient outcomes [331,332].   To facilitate administration, the 188Re-MS were suspended in a newly prepared HEC/Dex/CSB solution. The viscosity of this solution after sterilization was found to be 5.6 ± 0.2 cP at body temperature, or 11 times higher than water (0.5 ± 0.2 cP, measured at the same experimental conditions). Due to the high viscosity of the HEC/Dex/CSB solution, the 188Re-MS remain in suspension for longer. This was demonstrated experimentally by measuring the turbidity of a suspension of non-radioactive ~40 𝜇m MS over time (Figure 4.15). The optical density (𝑂𝐷) of the suspension of MS was measured at 450 nm every 30 s for 1 h using an ultraviolet/visible (UV/Vis) spectrophotometer (DU 800, Beckman Coulter; Brea, US). By suspending the 188Re-MS in the HEC/Dex/CSB solution, it is expected to reduce the settling rate of the 188Re-MS in the lumen of Catheter II, which simply acts as a reservoir for the 188Re-MS right before administration.         102   Figure 4.15. Turbidimetric Analysis of ~40 𝝁m Microspheres. The analysis was performed with non-radioactive MS from the same batch used to prepare the 188Re-MS (41.8 ± 6.0 𝜇m, 𝐶𝑉 = 14.4%). The MS were suspended at a concentration of 2 mg mL-1 in either saline or HEC/Dex/CSB solution (prepared with 1.75% w/v HEC, 5% w/v dextrose, and 0.01% w/v Chicago Sky Blue 6B). The MS remained in suspension for almost 10 min in HEC/Dex/CSB solution, but they started to precipitate in saline almost immediately (<2 min). Each data point is the mean of two independent measurements.    The osmolality of the parenteral formulation is important to avoid hemolysis (or lysis) and crenation (or shrinkage) of the red blood cells. Basically, hemolysis occurs in hypotonic formulations (i.e., <150 mOsm kg-1) and crenation occurs in hypertonic formulations (i.e., >600 mOsm kg-1). The osmolality is also important to reduce pain at the site of injection. The general recommendation is to work with isotonic formulations, where the osmolality is similar to plasma (i.e., 285-295 mOsm kg-1) [333]. The osmolality of the HEC/Dex/CSB solution was determined to be 289.5 ± 2.1 mOsm kg-1 and thus falls within the optimal range for utilization in vivo.   4.3.2.2 Development of a Hepatocellular Carcinoma-Bearing Animal Model   An important contribution of this pilot study is the development of an HCC-bearing animal model in male Sprague-Dawley rats. Currently, the gold standard in preclinical studies for radioembolization is the use of ~5 kg female New Zealand white rabbits bearing VX2 hepatic tumors. VX2 is a virus-induced papilloma which develops into an aggressive carcinoma in various organs of the rabbits (e.g., liver, rectum, lungs, and kidneys) [334]. The advantages of the VX2-bearing rabbit model are: 1) the blood supply to the VX2 hepatic tumors is almost entirely from the hepatic artery, 2) the hepatic artery is large enough to use a clinical catheter, and 3) the tumor achieves a size of up to 2 g in 2 weeks [335-337]. 103  This model, nonetheless, has faced criticism for a number of reasons. For instance, the implantation of VX2 hepatic tumors into the rabbits can be very challenging. A donor rabbit is typically required to grow an initial VX2 hepatic tumor. After excision from the donor rabbit, small sections of the tumor are implanted in the liver of the rest of the rabbits following a laparotomy [338,339]. In addition to the hepatic intra-arterial catheterization, which also involves a laparotomy, the rabbits are thus subjected to a complicated and costly surgical procedure a couple of weeks before radioembolization in order to grow the tumors. Furthermore, most preclinical equipment can only fit smaller animals, usually just mice and rats.  To overcome these shortcomings, a Novikoff-bearing rat model was successfully developed by inoculating a suspension of N1-S1 cells in MatrigelTM directly into the liver of the rats using ultrasound guidance. As a proof of concept, only two rats (R01 and R02, as per stated in Table 4.1) were subjected to this minimally invasive procedure, which resulted in the formation of a single tumor at the site of injection 2 weeks post-inoculation of the cells. To have an additional supply of cells in case of contamination, the initial culture of cells (referred to as C01) was split into two subcultures (referred to as C02 and C03). Both C02 and C03 were maintained in exponential growth phase until completion of the study.   In exponential growth phase, the number of cells doubles at an exponential rate which is dependent upon the cell line. This phase is characterized by cells dividing with genotypic and phenotypic stabilities. In stationary growth phase, nevertheless, the growth and the death rates are equal, or in other words, proliferation slows and eventually stops. Moreover, the cells might show variability in gene expression, leading to phenotypic changes [340-342]. Hence, only cells in exponential growth phase can proliferate and ultimately form a tumor in vivo. The preparation of the doses when the cells are in exponential growth phase also reduces inter-animal variation in tumor growth related to gene expression variability in the cells.  To ensure that C02 and C03 were in exponential growth phase, the population doubling level (𝑃𝐷𝐿) was calculated for the entire duration of the study. The 𝑃𝐷𝐿 is a measure of the age of the subcultures. Basically, it refers to the total number of times the cells in the population have doubled since the initial culture [343,344]. The 𝑃𝐷𝐿 was computed using the following equation:  𝑃𝐷𝐿 = 10.3 log𝑇𝑜𝑡𝑎𝑙 𝑉𝑖𝑎𝑏𝑙𝑒 𝐶𝑒𝑙𝑙𝑠 𝑎𝑡 𝐻𝑎𝑟𝑣𝑒𝑠𝑡𝑇𝑜𝑡𝑎𝑙 𝑉𝑖𝑎𝑏𝑙𝑒 𝐶𝑒𝑙𝑙𝑠 𝑎𝑡 𝑆𝑒𝑒𝑑     (Equation 4.1)  104  The variation in 𝑃𝐷𝐿 over time is depicted in Figure 4.16, where it is evident that neither C02 nor C03 reached a plateau in growth. A linear regression analysis yielded the following two equations for C02 and C03, respectively: Cumulative 𝑃𝐷𝐿 = 1.01t + 2.60 and Cumulative 𝑃𝐷𝐿 = 0.89t + 0.99. The coefficient of determination (𝑅2) was ~1 in both cases. The slopes of the linear equations correspond to the growth rates of the subcultures. Differences in the growth rates may occur, for example, if the cells are deprived of nutrients. The fact that the values are very similar implies that the subcultures maintained their integrity and thus can be employed interchangeably. For both R01 and R02, though, the doses of cells in MatrigelTM were prepared using C02.    Figure 4.16. Population Doubling Level of Subcultures of N1-S1 Cells. The cells were cultured at 37 °C and 5% CO2 in IMDM supplemented with 10% w/v FBS and 1% w/v Pen-Strep. The number of viable cells was counted using a hemocytomer grid following incubation of the samples with Trypan Blue. The graph shows that both C02 and C03 were maintained in exponential growth phase during the study.   The ultrasound-guided inoculation of cells was successfully performed in the rats. Each procedure can be easily completed in <30 min, from mounting the rat on the operating table to withdrawing the needle after inoculation of the cells. None of the rats experienced any complications during the procedure and recovered in the first 15 min following the injection.   Figure 4.17 depicts sonograms of the different stages of the procedure. The needle can be visualized on the screen and thus the exact site of injection is known. For instance, it is possible to determine how far from the skin the injection is performed. A minimum of 3 to 4 mm in depth is recommended to avoid abnormal tissue growth near the surface of the liver. A few seconds after inoculation of the cells, MatrigelTM begins to polymerize, resulting in a focus of cells at the 105  site of injection which is easily distinguished under ultrasound guidance. Ultimately, this provided evidence of the attachment of the cells to the tissue.       Figure 4.17. Stages of Ultrasound-Guided Inoculation of N1-S1 Cells into the Liver. The sonograms were acquired with the Vevo® 2100 Digital Imaging Platform (Visual Sonics; Toronto, Canada). The stages of the procedure are: (A) visualization of the liver, (B) perforation of the skin and the abdominal wall muscles followed by insertion of the needle into the liver, (C) injection of 1x106 cells suspended in 50 𝜇L of MatrigelTM, and (D) detection of a focus of cells after polymerization of the MatrigelTM (delineated in red). The images correspond to R01. The injection was performed on the left lobe.   4.3.2.3 Radioembolization and Quantitative SPECT Imaging   Due to the lack of equipment to monitor tumor progression in vivo, the presence of tumors in R01 and R02 was corroborated during necropsy. Indeed, post mortem examination revealed that both rats developed a single tumor in the left lateral lobe. R01 was euthanized 48 h post-radioembolization, whereas R02 was euthanized 2 weeks post-radioembolization. The tumors weighed 0.5 g. No tumors were found in any other organs.  The set of catheters described in Section 4.2.2.4.1 facilitated the administration of more than 95% of the 188Re-MS with minimal losses during the process. No significant losses of 188Re-MS occurred in the catheters, needles, and/or syringes. In fact, losses were only noticed in the pipette tip used to suspend the 188Re-MS in the HEC/Dex/CSB solution. The activity in all these tools, including the vial with the stock of 188Re-MS, was less than 0.7 MBq. The suspension of 188Re-MS was easily injected once Catheter I was secured at the injection site. The injection was performed in a slow and controlled fashion, requiring no more than 2 min for completion with no difficulties during administration. The set of catheters was flushed with 150 𝜇L of saline. The 106  administration device was found to be reliable, effective, and most important, safe. At no time was anyone exposed to the 𝛽− and 𝛾 emissions of 188Re. The fact that one of the excipients of the parenteral formulation was a dye, Chicago Sky Blue 6B, helped to monitor the flow of 188Re-MS in the catheters.  Due to the efficiency of administration, this strategy could be easily utilized in other preclinical studies involving the use of MS, either non-radioactive or radioactive. As illustrated in Figure 4.18, no extrahepatic deposition of 188Re-MS was observed in the SPECT/CT images. These uniformly-sized MS thus exhibit an improved embolic effect over 90Y-MS, where it has been reported that between 1 and 67% escape the liver and travel to the lungs via arteriovenous shunts (i.e., direct connections between the hepatic arterial system and the pulmonary venous system) [345]. As a matter of fact, shunting is one of the most problematic features of TheraSphere® and SIR-Spheres®, potentially leading to the development of radiation-induced pneumonitis [346,347]. R01 and R03 received an injection of 44.5 and 55.5 MBq of 188Re-MS, respectively, and are thus comparable to each other (the activity is similar). R02 and R04 received an injection of 12.6 and 11.5 MBq of 188Re-MS and are thus also comparable to each other. Generally, in the HCC-bearing rats (i.e., R01 and R02), the 188Re-MS accumulated predominantly in the tumor, whereas in the healthy rats (i.e., R03 and R04), the 188Re-MS distributed evenly throughout the liver.    Figure 4.18. Biodistribution of 188Re-Labeled Microspheres. Representative SPECT/CT images acquired 1 h-post radioembolization of (A) R01 (HCC/+; 47,700 188Re-MS, 44.4 MBq), (B) R02 (HCC/+; 47,700 188Re-MS, 12.6 MBq), (C) R03 (HCC/-; 71,600 188Re-MS, 55.5 MBq), and (D) R04 (HCC/-; 47,700 188Re-MS, 11.5 MBq). For R01 and R02, the tumor, as identified during necropsy, is delineated in red.  107  For the two rats in Cohort A, the 𝑇/𝐻 as well as the radiation doses to the tumor and the healthy liver tissue are summarized in Table 4.2. The radiation dose to the tumor was found to be three to six times larger than to the healthy liver tissue. The healthy liver tissue was exposed to less than 20 Gy in both cases, which is below the safety threshold of 70 Gy commonly employed in the clinical setting [348]. Therefore, not only were the 188Re-MS successfully administered into the liver, but they were also avidly taken up by the tumor. After radioembolization with 90Y-MS, 20 to 70% of the patients experience a number of adverse effects, such as abdominal pain/discomfort, cachexia, fever, nausea, and emesis. These adverse effects are collectively known as post-radioembolization syndrome and they are believed to be caused by deposition of 90Y-MS outside the tumor [349,350]. This pilot study has shown that quasi-monosized ~40 𝜇m 188Re-MS are preferentially delivered to the tumor while sparing most of the healthy liver tissue. As a result of their narrow size distribution, 188Re-MS could potentially reduce post-radioembolization syndrome incidence and severity. However, a long-term study with a larger sample size is required to determine if such adverse effects may occur after radioembolization with 188Re-MS.   Table 4.2. Radiation Doses to the Tumor and the Healthy Liver Tissue. Cohort A: HCC/+. Cohort B: HCC/-. The dosimetry calculations were performed with Monte Carlo using the dose kernel convolution method. Only SPECT images acquired 1 h post-radioembolization were used for the analysis.   Cohort ID Animal  ID Tumor  Mass (g) Liver  Mass (g) 𝑻/𝑯 Radiation Dose (Gy) Tumor Healthy  Liver Tissue  Cohort A R01 0.5 15 6.3 104 19.6 R02 0.5 11 3.1 22 7.6 Cohort B R03 N/A 14 N/A N/A 28.5 R04 N/A 15 N/A N/A 5.8   Due to the small blood vessels of the rats, the hepatic intra-arterial catheterization is a very challenging procedure which requires extensive surgical expertise. Before conducting a large-scale study, the outcomes of the surgery need improvement. The rats, for instance, need sufficient time to recover. To evaluate changes in the biodistribution of the 188Re-MS over time, the rats were imaged by SPECT/CT 1, 24, 48, and 72 h post-radioembolization. Only one of the rats, R02, recovered from all these scans and was euthanized 2 weeks post-radioembolization without any complications. The health of the remaining rats, nevertheless, deteriorated as additional SPECT/CT scans were performed. They were all euthanized a few days after the 108  surgery. Figure 4.19 shows that there are no changes in the biodistribution of 188Re-MS over time. Hence, multiple SPECT/CT scans can be omitted in future studies, which will potentially help to speed up the recovery of the rats.     Figure 4.19. Effect of Time on the Biodistribution 188Re-Labeled Microspheres. Representative SPECT/CT images of (A) R01 ([a] 1 h post-radioembolization and [b] 48 h post-radioembolization). (B) R03 ([c] 1 h post-radioembolization and [d] 72 h post-radioembolization). For R01, the tumor, as identified during necropsy, is delineated in red.   4.4 Conclusions  The flow focusing method, a microfluidic technology, was employed to generate size-defined Re-MS and 188Re-MS with a narrow size distribution: ~15 𝜇m Re-MS (𝐶𝑉 = 18.8%) and ~40 𝜇m 188Re-MS (𝐶𝑉 = 14.5%).  For the first time, intrinsically radiopaque polymer-based MS doped with rhenium were produced with a custom-synthesized polymer functionalized with rhenium. While uniformly-sized Re-MS were successfully developed, the proof of concept study described in this chapter was limited by the content of rhenium in the Re-MS. Upon increasing the content of rhenium, these Re-MS could be used as the first ever fully biodegradable and biocompatible embolic MS visible under X-rays. Based on the specific application, their size could be easily tuned by varying operational parameters of the flow focusing method (i.e., CP and DP flow rates).  Furthermore, MS were produced with a biodegradable and biocompatible polymer tailored with a chelator for 188Re binding on one end of the polymer chain. These MS were then labeled 109  with 188Re, a mixed 𝛽− and 𝛾 emitter. These 188Re-MS are truly theranostic: the short-range  𝛽− particles have a high cell killing power, whereas the 𝛾 photons are imageable by SPECT. The application of these 188Re-MS in radioembolization was assessed in a pilot study using an HCC-bearing rat model. This study demonstrated that quasi-monosized ~40 𝜇m MS ensure a complete embolization of the liver, with no shunting to the lungs. Additionally, it was shown that 188Re-MS are easily tracked by SPECT. In the clinical setting, this has the potential of allowing for quantitative imaging and accurate patient-specific dosimetry, which is lacking with currently used 90Y-MS.    110  Chapter 5 Conclusions  This dissertation has enhanced our understanding of the properties that make non-radioactive and radioactive rhenium useful in medical imaging. The following sections present an overall analysis and integration of the work described in this dissertation, including a discussion of the strengths, limitations, and future directions of this research.   5.1 Significance and Contribution of the Research  We found that rhenium attenuates X-rays significantly. This property was successfully exploited in the development of medical devices with potential applications in X-ray imaging. The medical devices include the production of an electrospun rhenium-doped scaffold and a sample of uniformly-sized Re-MS.  Regarding rhenium’s radioisotopes, 188Re is particularly interesting for applications in medical imaging because it decays by emission of both 𝛽− particles (which have a high cell killing power) and 𝛾 photons (which are “imageable” by SPECT). Using the same technology to prepare Re-MS, theranostic uniformly-sized 188Re-MS were produced. This dissertation explored specific applications for all of these medical devices.  To guide the research described in this dissertation, three aims were initially identified. The first aim was to evaluate the potential of rhenium in X-ray imaging by assessing image quality and absorbed dose. To evaluate image quality, the 𝐶𝑁𝑅s of rhenium- and iodine-based solutions were calculated in images acquired between 50 and 220 kVp. To estimate absorbed dose, the depth absorbed dose profiles from 50 to 120 kVp were predicted for a simulated 30 x 30 x 20 cm Solid Water® phantom with Monte Carlo. Our study showed that rhenium has a higher 𝐶𝑁𝑅 than iodine in high-kVp scans. Using Monte Carlo, we found that the average absorbed dose is lower in high-kVp scans than in low-kVp scans when images with similar quality and comparable noise are theoretically produced. High-kVp scans are commonly used for imaging average-sized and large patients, but noisy images are often acquired due to the suboptimal X-ray attenuation of iodine. In the clinical setting, image quality is often improved by increasing the exposure time, but this causes a higher than desired absorbed dose. Our study demonstrated that the utilization 111  of a rhenium in high-kVp scans has the potential of preserving image quality while minimizing radiation dose.  The second aim was to incorporate a radiopaque component onto otherwise radiolucent catheters by coating them with an electrospun scaffold doped with a rhenium complex. To achieve this aim, the first-ever radiopaque and fully biodegradable rhenium-doped scaffold was produced by electrospinning. The scaffold was made of PCL mixed with a rhenium complex. After applying heat, small samples of the scaffold were successfully adhered to the surface of catheters, forming a thin, uniform, and strongly radiopaque coating. Our study showed that the catheters became visible in images acquired by 𝜇CT. Making catheters radiopaque might be helpful for physicians in placing them inside the body more rapidly and precisely, potentially even with minimal or without the administration of I-XCAs.   Finally, the third aim was to prepare, characterize, and image uniformly-sized Re-MS and 188Re-MS. Using the flow focusing method, quasi-monosized ~15 𝜇m Re-MS and ~40 𝜇m 188Re-MS with a spherical shape and a smooth surface were successfully produced. The MS were made of a mixture of polymers: PLA, PEG-PCL, and a custom-synthesized polymer for binding of non-radioactive and radioactive rhenium. A preliminary evaluation of the applications of Re-MS in embolotherapy and 188Re-MS in radioembolization was included in this dissertation.   5.2 Strengths and Limitations of the Research  This dissertation summarizes one of the most comprehensive studies about the use of rhenium in X-ray imaging, including suggested concentrations to achieve a significant increase in contrast.  Melting our rhenium-doped scaffold onto the surface of catheters is a straightforward strategy to add a radiopaque component to catheters on a small scale. Due to its simplicity, this approach could be particularly valuable in research and preclinical applications. Our research showed that a rhenium/polymer matrix could be utilized as a source of contrast, but a cost-effective strategy to mass produce radiopaque catheters is needed for clinical applications.  Using a custom-synthesized rhenium-doped polymer, we showed that the production of ~15 𝜇m Re-MS by flow focusing is feasible. Moreover, our Re-MS have a very narrow size distribution, which is a desired characteristic for in vivo applications because the behavior of the MS is more predictable. We envision the use of very similar Re-MS as the first-ever intrinsically 112  radiopaque and fully biodegradable embolic MS, but further efforts should be made to increase the concentration of rhenium in the formulation and the size of the MS.   We demonstrated that larger radioactive MS can also be produced by flow focusing. In a small-scale pilot study in an HCC-bearing rat model, newly developed ~40 𝜇m 188Re-MS showed potential as an “imageable” radioembolization agent. We successfully determined the exact location of the 188Re-MS following administration using SPECT and calculated the radiation doses to the tumor and the healthy liver tissue using Monte Carlo. Based on the results of our pilot study, our uniformly-sized 188Re-MS are safe in regard to their biodistribution, achieving concentrations three to six times higher in the tumor than in the healthy liver tissue. This suggests that this novel radiopharmaceutical might be useful in improving the localization of the MS within the tumor. Furthermore, our 188Re-MS would most likely allow the design of treatment plans as well as the generation of 3D dosimetry maps for each patient.   5.3 Future Directions  Taking into account the now well-documented X-ray attenuation of rhenium, an interesting field to examine next is the development of very small metallic rhenium beads with dimensions at the nanoscale, or rhenium NPs (i.e., Re-NPs). Similar to the work conducted with other high-𝑍 elements (e.g., gold, bismuth, tantalum, ytterbium, and gadolinium), these Re-NPs could be investigated for utilization as an iodine-free NP-based XCA. An interesting possibility of incorporating a radiopaque component into catheters is the automatization of the coating step. Such an automatization might be challenging with electrospun scaffolds. Instead, a rhenium/polymer matrix could be spray coated onto catheters and/or a rhenium/polymer matrix could be fabricated or incorporated into specific sections of the catheters (e.g., during polymer extrusion). The second approach would be optimal in cases where there might be a change in the size of the catheters during use (e.g., when the balloon of a dilatation catheter is expanded).  Regarding our Re-MS, the following parameters require optimization: the concentration of rhenium in the formulation and the size of the MS. To increase the concentration of rhenium in the formulation, a strongly radiopaque rhenium-doped polymer should be synthesized. Each polymer chain in our current rhenium-doped polymer can only bind a single atom of rhenium. A work in progress is the synthesis of a chelomer where each polymer chain contains more than 113  one rhenium binding group. To increase the size of the MS, larger microfluidic chips must be made or a different microfluidic technology must be utilized. Due to the size of the microchannels in our current flow focusing microchip, only MS with a diameter up to 50 𝜇m can be reliably produced. Recently, our laboratory has made significant progress on the development of a micro co-flowing device to produce 100 to 700 𝜇m polymer-based MS with a 𝐶𝑉 of less than 10% (submitted manuscript by Zeynab Nosrati, a Ph.D. student in the Hafeli Laboratory). Once a strongly radiopaque rhenium-doped polymer is readily available, it will be worth investigating the feasibility of using this new technology to produce larger Re-MS, suitable for embolotherapy. Such Re-MS would show exactly where they end up in the target tissue, with relatively high angiographic resolution. Additionally, it would be possible to infuse more MS into areas that received lower concentrations or purposely increase the concentrations in tumor areas where cancer recurrence is known to be high.  Our radioembolization study with 188Re-MS showed promising results. To move this project forward, it will be interesting to compare the biodistribution of 188Re-MS with different sizes, specifically ~30, ~40, and ~50 𝜇m (all of which are easily produced by the flow focusing described in this dissertation). Potential differences in the degree of shunting to the lungs will provide valuable of information of the importance of size and size distribution in radioembolization. Long-term studies with a larger sample size are required in order to determine the incidence of adverse effects following the administration of 188Re-MS. In this study, the therapeutic efficacy of 188Re-MS should be investigated as well by monitoring changes in the size of the tumors. Over the last few years, intensity modulated radiotherapy (IMRT) has rapidly become the state of the art conformal radiotherapy (a form of external beam radiotherapy) modality of choice world-wide. IMRT shapes a high-energy X-ray beam to closely fit tumor areas, thereby minimizing damage to surrounding healthy tissue [351-353]. The outcomes of using IMRT following radioembolization with 188Re-MS could also be studied to gain in-depth knowledge about possible benefits of this combination approach in the treatment of HCC.  The chelomer used to prepare our 188Re-MS also binds 99mTc, which has been previously reported by Häfeli et al. [306]. The fact that the same MS can be labeled with either 188Re or 99mTc offers very exciting opportunities. In Section 4.1.1.1, it was mentioned that a hepatic perfusion study with 99mTc-MAA is conducted prior to radioembolization with 90Y-MS to assess the degree of portal shunting to the lungs. This study requires the administration of 99mTc-MAA 114  into the patient via a hepatic intra-arterial catheterization (i.e., simulating exactly the radioembolization procedure with 90Y-MS). The degree of portal shunting to the lungs is determined by imaging 99mTc-MAA, typically with a planar 2D scan. If more than 10% of the activity injected is found in the lungs, radioembolization is contraindicated. Nevertheless, 99mTc-MAA does not have the same size, size distribution, and density as 90Y-MS. Therefore, the biodistribution of 99mTc-MAA is not identical to 90Y-MS, making 99mTc-MAA less than desirable for this application [294,295]. We propose to label our MS with 99mTc (i.e., 99mTc-MS) and evaluate their potential utilization as a replacement of 99mTc-MAA. These 99mTc-MS, which will be identical in size, size distribution, and density to our 188Re-MS, could be imaged by SPECT/CT for pre-radioembolization lung shunting assessment and treatment planning, including image-based personalized dosimetry calculations. Then, for radioembolization, identical MS could be labeled with 188Re in patient-specific activity concentrations, determined from the 99mTc-MS scans. Moreover, the number and the size of MS could be easily adjusted for each patient.   The research presented in this dissertation has demonstrated the immense potential of non-radioactive and radioactive rhenium for utilization in medical imaging. Our findings provide a strong foundation to guide other researchers interested in further investigating rhenium. This dissertation opens up many possibilities, with several applications of rhenium yet to be explored.       115  References  [1] Moustapha ME, Ehrhardt GJ, Smith CJ, et al. (2006). Preparation of Cyclotron-Produced 186Re and Comparison with Reactor-Produced 186Re and Generator-Produced 188Re for the Labeling of Bombesin. Nuclear Medicine and Biology. 33: 81-89. [2] Emsley J (2001). Nature's Building Blocks: An A-Z Guide to the Elements. Oxford University Press Inc. New York, USA.  [3] Sutulov A (1976). Molybdenum and Rhenium. 1778 - 1977. University of Concepción; Concepción, Chile. [4] Esquinas PL, Rodríguez-Rodríguez C, De La Vega JC, et al. (2017). 188Re Image Performance Assessment Using Small Animal Multi-Pinhole SPECT/PET/CT System. Physica Medica: European Journal of Medical Physics. 33: 26-37. [5] Boschi A, Uccelli L, Pasquali M, et al. (2014). 188W/188Re Generator System and its Therapeutic Applications. Journal of Chemistry. 2014:529406. [6] Pillai MRA, Dash A, and Knapp FF (2012). Rhenium-188: Availability from the 188W/188Re Generator and Status of Current Applications. Current Radiopharmaceuticals. 5: 228-243. [7] de Goeij JJM and Bonardi ML (2005). How Do We Define the Concepts Specific Activity, Radioactive Concentration, Carrier, Carrier-Free and No-Carrier-Added? Journal of Radioanalytical and Nuclear Chemistry. 263: 13-18. [8] Dilworth JR and Parrott SJ (1998). The Biomedical Chemistry of Technetium and Rhenium. Chemical Society Reviews. 27: 43-55. [9] Deutsch E, Libson K, Vanderheyden JL, et al. (1986). The Chemistry of Rhenium and Technetium as Related to the Use of Isotopes of these Elements in Therapeutic and Diagnostic Nuclear Medicine. International Journal of Radiation Applications and Instrumentation. Part B, Nuclear Medicine and Biology. 13: 465-477. [10] Iikuni S, Ono M, Watanabe H, et al. (2016). Imaging of Cerebral Amyloid Angiopathy with Bivalent 99mTc-Hydroxamamide Complexes. Scientific Reports. 6: 25990. [11] Iikuni S, Ono M, Watanabe H, et al. (2016). Novel Bivalent 99mTc-Complex with N-Methyl-Substituted Hydroxamamide as Probe for Imaging of Cerebral Amyloid Angiopathy. PLoS ONE. 11: e0163969. [12] Zhang Q, Guan Y, Liu S, et al. (2016). Synthesis and Biological Evaluation of a New Nitroimidazole-99mTc-Complex for Imaging of Hypoxia in Mice Model. Medical Science Monitor. 22: 3778-3791. [13] Liu S (2004). The Role of Coordination Chemistry in the Development of Target-Specific Radiopharmaceuticals. Chemical Society Reviews. 33: 445-461. [14] Zhuang ZP, Kung M-P, Mu, et al. (1999). 99mTc-Labeled MIBG Derivatives:  Novel 99mTc Complexes as Myocardial Imaging Agents for Sympathetic Neurons. Bioconjugate Chemistry. 10: 159-168. [15] Arnold RW, Subramanian G, McAfee JG, et al. (1975). Comparison of 99mTc Complexes for Renal Imaging. Journal of Nuclear Medicine. 16: 357-367. [16] Seridi A, Wolff M, Boulay A, et al. (2011). Rhenium(I) and Technetium(I) Complexes of a Novel Pyridyltriazole-Based Ligand Containing an Arylpiperazine Pharmacophore: Synthesis, Crystal Structures, Computational Studies and Radiochemistry. Inorganic Chemistry Communications. 14: 238-242. [17] Saatchi K and Häfeli UO (2007). One-Pot Syntheses, Coordination, and Characterization of Application-Specific Biodegradable Ligand-Polymers. Dalton Transactions. 4439-4445. [18] Stephenson KA, Banerjee SR, Besanger T, et al. (2004). Bridging the Gap between in Vitro and in Vivo Imaging:  Isostructural Re and 99mTc Complexes for Correlating Fluorescence and Radioimaging Studies. Journal of the American Chemical Society. 126: 8598-8599. 116  [19] Gali H, Hoffman TJ, Sieckman GL, et al. (2001). Synthesis, Characterization, and Labeling with 99mTc/188Re of Peptide Conjugates Containing a Dithia-Bisphosphine Chelating Agent. Bioconjugate Chemistry. 12: 354-363. [20] Kaur T, Lee WZ, and Ravikanth M (2016). Rhenium(I) Tricarbonyl Complexes of Meso-Tetraaryl-21,23-Diheteroporphyrins. Inorganic Chemistry. 55: 5305-5311. [21] Bartholoma MD, Vortherms AR, Hillier S, et al. (2011). Synthesis, Cytotoxicity and Cellular Uptake Studies of N3 Functionalized Re(CO)3 Thymidine Complexes. Dalton Transactions. 40: 6216-6225. [22] Binkley SL, Barone NV, Underwood AC, et al. (2010). The Synthesis and Toxicity of Tripodal Tricarbonyl Rhenium Complexes as Radiopharmaceutical Models. Journal of Inorganic Biochemistry. 104: 632-638. [23] Fernandez-Moreira V, Thorp-Greenwood FL, Amoroso AJ, et al. (2010). Uptake and Localisation of Rhenium fac-Tricarbonyl Polypyridyls in Fluorescent Cell Imaging Experiments. Organic & Biomolecular Chemistry. 8: 3888-3901. [24] Choi SJ, Brylev KA, Xu JZ, et al. (2008). Cellular Uptake and Cytotoxicity of Octahedral Rhenium Cluster Complexes. Journal of Inorganic Biochemistry. 102: 1991-1996. [25] Fuks L, Gniazdowska E, Mieczkowski J, et al. (2008). Structural Features of Tricarbonyl(N-Methyl-2-Pyridinecarboxyamide)Chloro-Rhenium(I)-Potential Precursor of Radiopharmaceuticals. Polyhedron. 27: 1353-1360. [26] Vogler A and Kunkely H (2000). Excited State Properties of Organometallic Compounds of Rhenium in High and Low Oxidation States. Coordination Chemistry Reviews. 200–202: 991-1008. [27] Abedi S (2015). Therapeutic Applications and Dosimetry of Radiopharmaceuticals with the Help of Compartmental Analysis. In: Jaffray DA (ed). World Congress on Medical Physics and Biomedical Engineering; June 7-12, 2015, Toronto, Canada. Springer International Publishing. 674-678. [28] Kutzner J, Becker M, and Grimm W (1983). Zur Osteotropie Von Rhenium- und Yttrium-Komplexen. Nuklearmedizin. 22: 162-165. [29] Durbin PW (1960). Metabolic Characteristics within a Chemical Family. Health Physics. 2: 225-238. [30] Häfeli UO, Casillas S, Dietz DW, et al. (1999). Hepatic Tumor Radioembolization in a Rat Model Using Radioactive Rhenium (186Re/188Re) Glass Microspheres. International Journal of Radiation Oncology*Biology*Physics. 44: 189-199. [31] Huda W and Abrahams RB (2015). X-Ray-Based Medical Imaging and Resolution. American Journal of Roentgenology. 204: W393-W397. [32] Yaffe MJ and Rowlands JA (1997). X-Ray Detectors for Digital Radiography. Physics in Medicine and Biology. 42: 1-39.  [33] Cormode DP, Naha PC and Fayad ZA (2014). Nanoparticle Contrast Agents for Computed Tomography: A Focus on Micelles. Contrast Media & Molecular Imaging. 9: 37-52. [34] Garvey CJ and Hanlon R (2002). Computed Tomography in Clinical Practice. BMJ. 324: 1077-1080. [35] Hallouard F, Anton N, Choquet P, et al. (2010). Iodinated Blood Pool Contrast Media for Preclinical X-Ray Imaging Applications - A Review. Biomaterials. 31: 6249-6268. [36] Ritman EL (2004). Micro-Computed Tomography - Current Status and Developments. Annual Review of Biomedical Engineering. 6: 185-208. [37] Ritman EL (2011). Current Status of Developments and Applications of Micro-CT. Annual Review of Biomedical Engineering. 13: 531-552. [38] Yepes JF and Al-Sabbagh M (2015). Use of Cone-Beam Computed Tomography in Early Detection of Implant Failure. Dental Clinics of North America. 59: 41-56. [39] Venskutonis T, Plotino G, Juodzbalys G, et al. (2014). The Importance of Cone-Beam Computed Tomography in the Management of Endodontic Problems: A Review of the Literature. Journal of Endodontics. 40: 1895-1901. 117  [40] López FU, Kopper PMP, Cucco C, et al. (2014). Accuracy of Cone-Beam Computed Tomography and Periapical Radiography in Apical Periodontitis Diagnosis. Journal of Endodontics. 40: 2057-2060. [41] Sarment DP and Christensen AM (2014). The Use of Cone Beam Computed Tomography in Forensic Radiology. Journal of Forensic Radiology and Imaging. 2: 173-181. [42] Levine MS, Rubesin SE and Laufer I (2009). Barium Studies in Modern Radiology: Do They Have a Role? Radiology. 250: 18-22. [43] Ho LM, Nelson RC and DeLong DM (2007). Determining Contrast Medium Dose and Rate on Basis of Lean Body Weight: Does this Strategy Improve Patient-to-Patient Uniformity of Hepatic Enhancement During Multi-Detector Row CT? Radiology. 243: 431-437. [44] GE Healthcare Inc. (2010). OmnipaqueTM (Iohexol) Injection. Princeton, USA: 8. [45] Pan Y-N, Li A-J, Chen X-M, et al. (2016). Coronary Computed Tomography Angiography at Low Concentration of Contrast Agent and Low Tube Voltage in Patients with Obesity. Academic Radiology. 23: 438-445. [46] Lee SY, Rhee CM, Leung AM, et al. (2015). A Review: Radiographic Iodinated Contrast Media-Induced Thyroid Dysfunction. The Journal of Clinical Endocrinology and Metabolism. 100: 376-383. [47] Bottinor W, Polkampally P and Jovin I (2013). Adverse Reactions to Iodinated Contrast Media. International Journal of Angiology. 22: 149-154. [48] Andreucci M, Solomon R and Tasanarong A (2014). Side Effects of Radiographic Contrast Media: Pathogenesis, Risk Factors, and Prevention. BioMed Research International. 2014: 741018. [49] Lightfoot CB, Abraham RJ, Mammen T, et al. (2009). Survey of Radiologists' Knowledge Regarding the Management of Severe Contrast Material-Induced Allergic Reactions. Radiology. 251: 691-696. [50] Pumphrey (2000). Lessons for Management of Anaphylaxis from a Study of Fatal Reactions. Clinical & Experimental Allergy. 30: 1144-1150. [51] Marin D, Nelson RC, Rubin GD, et al. (2011). Body CT: Technical Advances for Improving Safety. American Journal of Roentgenology. 197: 33-41. [52] Nakayama Y, Awai K, Funama Y, et al. (2006). Lower Tube Voltage Reduces Contrast Material and Radiation Doses on 16-MDCT Aortography. American Journal of Roentgenology. 187: W490-W497. [53] Fouras A, Kitchen MJ, Dubsky S, et al. (2009). The Past, Present, and Future of X-Ray Technology for in vivo Imaging of Function and Form. Journal of Applied Physics. 105: 102009. [54] Huda W, Lieberman KA, Chang J, et al. (2004). Patient Size and X-Ray Technique Factors in Head Computed Tomography Examinations. I. Radiation Doses. Medical Physics. 31: 588-594. [55] Huda W, Lieberman KA, Chang J, et al. (2004). Patient Size and X-Ray Technique Factors in Head Computed Tomography Examinations. II. Image Quality. Medical Physics. 31: 595-601. [56] Ogden K and Huda W (2004). Patient Size and X-Ray Transmission in Body CTt. Health Physics. 84: 397-405. [57] Khan FM (1994). The Physics of Radiation Therapy. 2nd Edition. Lippincott Williams & Wilkins. Philadelphia, USA. [58] De La Vega JC and Häfeli UO (2015). Utilization of Nanoparticles as X-Ray Contrast Agents for Diagnostic Imaging Applications. Contrast Media & Molecular Imaging. 10: 81-95. [59] Cecil RL (1940). Gold Therapy in Rheumatoid Arthritis. Transactions of the American Clinical and Climatological Association. 56: 187-198. [60] Naha PC, Chhour P and Cormode DP (2015). Systematic in Vitro Toxicological Screening of Gold Nanoparticles Designed for Nanomedicine Applications. Toxicology in Vitro. 29: 1445-1453. [61] Connor EE, Mwamuka J, Gole A, et al. (2005). Gold Nanoparticles are Taken Up by Human Cells but Do Not Cause Acute Cytotoxicity. Small. 1: 325-327. [62] Hainfeld JF, Slatkin DN, Focella TM, et al. (2006). Gold Nanoparticles: A New X-Ray Contrast Agent. British Journal of Radiology. 79: 248-253. 118  [63] Cheheltani R, Ezzibdeh RM, Chhour P, et al. (2016). Tunable, Biodegradable Gold Nanoparticles as Contrast Agents for Computed Tomography and Photoacoustic Imaging. Biomaterials. 102: 87-97. [64] Naha PC, Lau KC, Hsu JC, et al. (2016). Gold Silver Alloy Nanoparticles (GSAN): An Imaging Probe for Breast Cancer Screening with Dual-Energy Mammography or Computed Tomography. Nanoscale. 8: 13740-13754. [65] Jackson PA, Rahman WNWA, Wong CJ, et al. (2010). Potential Dependent Superiority of Gold Nanoparticles in Comparison to Iodinated Contrast Agents. European Journal of Radiology. 75: 104-109. [66] Dreaden EC, Alkilany AM, Huang X, et al. (2012). The Golden Age: Gold Nanoparticles for Biomedicine. Chemical Society Reviews. 41: 2740-2779. [67] Kim F, Connor S, Song H, et al. (2004). Platonic Gold Nanocrystals. Angewandte Chemie. 116: 3759-3763. [68] Sun Y and Xia Y (2002). Shape-Controlled Synthesis of Gold and Silver Nanoparticles. Science. 298: 2176-2179. [69] Sun H, Li H, Harvey H, et al. (1999). Interactions of Bismuth Complexes with Metallothionein(II). The Journal of Biological Chemistry. 274: 29094-29101. [70] Brown AL, Naha PC, Benavides-Montes V, et al. (2014). Synthesis, X-Ray Opacity, and Biological Compatibility of Ultra-High Payload Elemental Bismuth Nanoparticle X-Ray Contrast Agents. Chemistry of Materials. 26: 2266-2274. [71] Brown AL and Goforth AM (2012). pH-Dependent Synthesis and Stability of Aqueous, Elemental Bismuth Glyconanoparticle Colloids: Potentially Biocompatible X-Ray Contrast Agents. Chemistry of Materials. 24: 1599-1605. [72] Ai K, Liu Y, Liu J, et al. (2011). Large-Scale Synthesis of Bi2S3 Nanodots as a Contrast Agent for in Vivo X-Ray Computed Tomography Imaging. Advanced Materials. 23: 4886-4891. [73] Rabin O, Manuel Perez J, Grimm J, et al. (2006). An X-Ray Computed Tomography Imaging Agent Based on Long-Circulating Bismuth Sulphide Nanoparticles. Nature Materials. 5: 118-122. [74] Nadel JA, Wolfe WG, and Graf PD (1968). Powdered Tantalum as a Medium for Bronchography in Canine and Human Lungs. Investigative Radiology. 3: 229-238. [75] Bonitatibus JPJ, Torres AS, Goddard GD, et al. (2010). Synthesis, Characterization, and Computed Tomography Imaging of a Tantalum Oxide Nanoparticle Imaging Agent. Chemical Communications. 46: 8956-8958. [76] Bonitatibus PJ, Torres AS, Kandapallil B, et al. (2012). Preclinical Assessment of a Zwitterionic Tantalum Oxide Nanoparticle X-Ray Contrast Agent. ACS Nano. 6: 6650-6658. [77] Vancaeyzeele C, Ornatsky O, Baranov V, et al. (2007). Lanthanide-Containing Polymer Nanoparticles for Biological Tagging Applications:  Nonspecific Endocytosis and Cell Adhesion. Journal of the American Chemical Society. 129: 13653-13660. [78] Yan G-P, Zheng C-Y, Cao W, et al. (2003). Synthesis and Preliminary Evaluation of Gadolinium Complexes Containing Sulfonamide Groups as Potential MRI Contrast Agents. Radiography. 9: 35-41. [79] Zhou X, Yang L, Yan G, et al. (2012). Dopamine-Containing Gadolinium Complex as Magnetic Resonance Imaging Contrast Agent. Journal of Rare Earths. 30: 884-889. [80] Liu Z, Li Z, Liu J, et al. (2012). Long-Circulating Er3+-Doped Yb2O3 up-Conversion Nanoparticle as an in Vivo X-Ray CT Imaging Contrast Agent. Biomaterials. 33: 6748-6757. [81] Xing H, Bu W, Ren Q, et al. (2012). A NaYbF4: Tm3+ Nanoprobe for CT and NIR-to-NIR Fluorescent Bimodal Imaging. Biomaterials. 33: 5384-5393. [82] Krasilnikova AA, Soloviera AO, Ivanov AA, et al. (2017). A Comparative Study of Hydrophilic Phosphine Hexanuclear Rhenium Cluster Complexes' Toxicity. Toxicology Research. 6: 554-560. [83] Krasilnikova AA, Soloviera AO, Ivanov AA, et al. (2017). Comprehensive Study of Hexarhenium Cluster Complex Na4[{Re6Te8}(CN)6] - In Terms of a New Promising Luminiscent and X-Ray Contrast Agent. Nanomedicine: Nanotechnology, Biology, and Medicine. 13: 755-763. 119  [84] Krasilnikova AA, Soloviera AO, Trifonova KE, et al. (2016). Cellular Internalization and Morphological Analysis after Intravenous Injection of a Highly Hydrophilic Octahedral Rhenium Cluster Complex - A New Promising X-Ray Contrast Agent. Contrast Media & Molecular Imaging. 11: 459-466. [85] Krasilnikova AA, Shestopalov MA, Brylev KA, et al. (2015). Prospects of Molybdenum and Rhenium Octahedral Cluster Complexes as X-Ray Contrast Agents. Journal of Inorganic Biochemistry. 144: 13-17. [86] Patterson JR (2017). They Tamed the Photon: Early Days of Radiation Dosimetry. Australasian Physical & Engineering Sciences in Medicine. 1: 11-20. [87] Tubiana M (1996). Wilhelm Conrad Röntgen and the Discovery of X-Rays. Bulletin de l'Académie Nationale de Médecine. 180: 97-108. [88] Seibert JA (2004). X-Ray Imaging Physics for Nuclear Medicine Technologists. Part 1: Basic Principles of X-Ray Production. Journal of Nuclear Medicine Technology. 32: 139-147. [89] Bailey DL, Humm JL, Todd-Pokropek A, et al. (2014). Nuclear Medicine Physics: A Handbook for Teachers and Students. International Atomic Energy Agency. Vienna, Austria. [90] Seibert JA and Boone JM (2005). X-Ray Imaging Physics for Nuclear Medicine Technologists. Part 2: X-Ray Interactions and Image Formation. Journal of Nuclear Medicine Technology. 33: 3-18. [91] Koedooder K (1986). Dose Reduction by X-Ray Beam Filtration in Screen-Film Radiography: Experimental and Theoretical Aspects. University of Amsterdam. Amsterdamn, Netherlands. [92] Brosi P, Stuessi A, Verdun FR, et al. (2011). Copper Filtration in Pediatric Digital X-Ray Imaging: Its Impact on Image Quality and Dose. Radiological Physics and Technology. 4: 148-155. [93] Wong MD, Wu X and Liu H (2011). The Effects of X-Ray Beam Hardening on Detective Quantum Efficiency and Radiation Dose. Journal of X-Ray Science and Technology. 19: 509-519. [94] Brooks RA and Chiro GD (1976). Beam Hardening in X-Ray Reconstructive Tomography. Physics in Medicine and Biology. 21: 390-398. [95] Martin CJ (2007). Optimisation in General Radiography. Biomedical Imaging and Intervention Journal. 3: e18. [96] Martin CJ (2007). The Importance of Radiation Quality for Optimisation in Radiology. Biomedical Imaging and Intervention Journal. 3: e38. [97] Oliveira ML, Tosoni GM, Lindsey DH, et al. (2014). Assessment of CT Numbers in Limited and Medium Field-of-View Scans Taken Using Accuitomo 170 and Veraviewepocs 3De Cone-Beam Computed Tomography Scanners. Imaging Science in Dentistry. 44: 279-285. [98] Huda W and Abrahams RB (2015). Radiographic Techniques, Contrast, and Noise in X-Ray Imaging. American Journal of Roentgenology. 204: W126-W131. [99] Schwab F, Schleede S, Hahn D, et al. (2013). Comparison of Contrast-to-Noise Ratios of Transmission and Dark-Field Signal in Grating-Based X-Ray Imaging for Healthy Murine Lung Tissue. Zeitschrift für Medizinische Physik. 23: 236-242. [100] Diekmann F, Sommer A, Lawaczeck R, et al. (2007). Contrast-to-Noise Ratios of Different Elements in Digital Mammography: Evaluation of Their Potential as New Contrast Agents. Investigative Radiology. 42: 319-325. [101] Doyle P, Martin CJ, and Gentle D (2006). Application of Contrast-to-Noise Ratio in Optimizing Beam Quality for Digital Chest Radiography: Comparison of Experimental Measurements and Theoretical Simulations. Physics in Medicine and Biology. 51: 2953-2970. [102] Mah P, Reeves TW, and McDavid WD (2010). Deriving Hounsfield Units Using Grey Levels in Cone Beam Computed Tomography. Dentomaxillofacial Radiology. 39: 323-335. [103] Hall EJ (2000). Radiobiology for the Radiologist. 5th Edition. Lippincott Williams & Wilkins. Philadelphia, USA. 120  [104] Aliasgharzadeh A, Mihandoost E, Masoumbeigi M, et al. (2015). Measurement of Entrance Skin Dose and Calculation of Effective Dose for Common Diagnostic X-Ray Examinations in Kashan, Iran. Global Journal of Health Science. 7: 202-207. [105] Hufton AP (2002). Diagnostic Radiology: Patient Dosimetry. In: Martin CJ and Sutton DG (eds). Practical Radiation Protection in Health Care. Oxford University Press. Oxford, UK: 221-241. [106] Goldman LW (2007). Principles of CT: Radiation Dose and Image Quality. Journal of Nuclear Medicine Technology. 35: 213-225. [107] Sander T (2014). Air Kerma and Absorbed Dose Standards for Reference Dosimetry in Brachytherapy. The British Journal of Radiology. 87: 20140176. [108] FitzGerald PF, Colborn RE, Edic PM, et al. (2016). CT Image Contrast of High-Z Elements: Phantom Imaging Studies and Clinical Implications. Radiology. 278: 723-733. [109] Du LY, Umoh J, Nikolov HN, et al. (2007). A Quality Assurance Phantom for the Performance Evaluation of Volumetric Micro-CT Systems. Physics in Medicine and Biology. 52: 7087-7108. [110] Armour M, Ford E, Iordachita I, et al. (2009). CT Guidance Is Needed to Achieve Reproducible Positioning of the Mouse Head for Repeat Precision Cranial Irradiation. Radiation Research. 173: 119-123. [111] Mohammad M, Eric F, Iulian I, et al. (2009). Image-Guided Small Animal Radiation Research Platform: Calibration of Treatment Beam Alignment. Physics in Medicine and Biology. 54: 891-905. [112] Kagadis GC, Ford NL, Karnabatidis D, et al. (2016). Handbook of Small Animal Imaging: Preclinical Imaging, Therapy, and Applications. CRC Press. Taylor & Francis Group. New York, USA.  [113] Behiels G, Maes F, Vandermeulen D, et al. (2002). Retrospective Correction of the Heel Effect in Hand Radiographs. Medical Image Analysis. 6: 183-190. [114] Braun H, Kachelrieß M, Kyriakou Y, et al. (2009). The Influence of the Heel Effect on FD-CT Imaging with Shifted Detectors. In: Dössel O and Schlegel WC (eds). World Congress on Medical Physics and Biomedical Engineering; September 7-12, 2009, Munich, Germany. Springer. Heidelberg, Germany. 439-442. [115] do Nascimento MZ, Frère AF, and Germano F (2008). An Automatic Correction Method for the Heel Effect in Digitized Mammography Images. Journal of Digital Imaging. 21: 177-187. [116] Rana V, Gill K, Rudin S, et al. (2012). Significance of Including Field Non-Uniformities Such as the Heel Effect and Beam Scatter in the Determination of the Skin Dose Distribution During Interventional Fluoroscopic Procedures. Proceedings of SPIE - The International Society for Optical Engineering. 8313: 83131N. [117] Parsa A, Ibrahim N, Hassan B, et al. (2012). Reliability of Voxel Gray Values in Cone Beam Computed Tomography for Preoperative Implant Planning Assessment. International Journal of Oral & Maxillofacial Implants. 27: 1438-1442. [118] Valiyaparambil JV, Yamany I, Ortiz D, et al. (2012). Bone Quality Evaluation: Comparison of Cone Beam Computed Tomography and Subjective Surgical Assessment. International Journal of Oral & Maxillofacial Implants. 27: 1271-1277. [119] Nomura Y, Watanabe H, Honda E, et al. (2010). Reliability of Voxel Values from Cone-Beam Computed Tomography for Dental Use in Evaluating Bone Mineral Density. Clinical Oral Implants Research. 21: 558-562. [120] Kimpe T and Tuytschaever T (2007). Increasing the Number of Gray Shades in Medical Display Systems -  How Much Is Enough? Journal of Digital Imaging. 20: 422-432. [121] Li X, Morgan PS, Ashburner J, et al. (2016). The First Step for Neuroimaging Data Analysis: DICOM to NIFTI Conversion. Journal of Neuroscience Methods. 264: 47-56. [122] Kahn Jr CE, Carrino JA, Flynn MJ, et al. (2007). DICOM and Radiology: Past, Present, and Future. Journal of the American College of Radiology. 4: 652-657. 121  [123] Mildenberger P, Eichelberg M and Martin E (2002). Introduction to the DICOM Standard. European Radiology. 12: 920-927. [124] Barten PGJ (1992). Physical Model for the Contrast Sensitivity of the Human Eye. SPIE Proceedings. 1666: 57-72. [125] Gillam JE, Kitcher D, Beveridge TE, et al. (2009). K-Edge Subtraction Using an Energy-Resolving Position-Sensitive Detector. Nuclear Instruments and Methods in Physics Research; Section A: Accelerators, Spectrometers, Detectors and Associated Equipment. 604: 97-100. [126] National Physical Laboratory (NPL). Absorption of Photons: X-Ray Absorption Edges, Characteristic X-Ray Lines and Fluorescence Yields. http://www.kayelaby.npl.co.uk/atomic_and_nuclear_physics/4_2/4_2 _1.html [Accessed on February 8, 2017]. [127] Beeman WW and Friedman H (1939). The X-Ray K Absorption Edges of the Elements Fe (26) to Ge (32). Physical Review. 56: 392-405. [128] National Physical Laboratory (NPL). X-Ray Transition Energy: K-Edge Table. http://physics.nist.gov/cgi-bin/XrayTrans/search.pl?download=column&element=All&trans=Kedge&lower=8000&upper=200000&units=eV [Accessed on February 8, 2017].  [129] Boone JM and Seibert JA (1997). An Accurate Method for Computer-Generating Tungsten Anode X-Ray Spectra from 30 to 140 Kv. Medical Physics. 24: 1661-1670. [130] Siewerdsen JH, Waese AM, Moseley DJ, et al. (2004). Spektr: A Computational Tool for X-Ray Spectral Analysis and Imaging System Optimization. Medical Physics. 31: 3057-3067. [131] Shrimpton PC, Jones DG and Wall BF (1988). The Influence of Tube Filtration and Potential on Patient Dose During X-Ray Examinations. Physics in Medicine and Biology. 33: 1205-1212. [132] Goo HW (2012). CT Radiation Dose Optimization and Estimation: An Update for Radiologists. Korean Journal of Radiology. 13: 1-11. [133] Perlmutter N, Arthur RJ, Beluffi G, et al. (1998). The Quality Criteria for Diagnostic Radiographic Images in Paediatrics. Radiation Protection Dosimetry. 80: 45-48. [134] Kuramoto T, Ohshita M, Maeda K, et al. (2007). Sharpening of an Energy Band of Diagnostic X-Ray Spectrum with Metal Filters. In: Magjarevic R and Nagel JH (eds). World Congress on Medical Physics and Biomedical Engineering 2006; August 27-September 1, 2006, Seoul, Korea. Springer. Heidelberg, Germany: 1533-1536. [135] Carmichael JHE, Maccia C, Moores BM, et al. (1996). European Guidelines on Quality Criteria for Diagnostic Radiographic Images. In: Comission E (ed). Europa Server. Luxembourg City, Luxembourg: 98. [136] Kohn MM, Moores BM, Schibilla H, et al. (1996). European Guidelines on Quality Criteria for Diagnostic Radiographic Images in Paediatrics. In: Comission E (ed). Europa Server. Luxembourg City, Luxembourg: 71. [137] Ledezma CJ and Wintermark M (2009). Multi-Modal CTt in Stroke Imaging: New Concepts. Radiologic Clinics of North America. 47: 109-116. [138] Duan X, Wang J, Yu L, et al. (2011). CT Scanner X-Ray Spectrum Estimation from Transmission Measurements. Medical Physics. 38: 993-997. [139] Yeh BM, Shepherd JA, Wang ZJ, et al. (2009). Dual Energy and Low kVp CT in the Abdomen. American Journal of Roentgenology. 193: 47-54. [140] Kaza RK, Platt JF, Cohan RH, et al. (2012). Dual-Energy CT with Single- and Dual-Source Scanners: Current Applications in Evaluating the Genitourinary Tract. Radiographics. 32: 353-369. [141] Barnes GT, Sones RA, Tesic MM, et al. (1985). Detector for Dual-Energy Digital Radiography. Radiology. 156: 537-540. [142] Al Khalifah K and Brindhaban (2004). A Comparison Between Conventional Radiography and Digital Radiography for Various kVp and mAs Settings Using a Pelvic Phantom. Radiography. 10: 119-125. 122  [143] Penner-Hahn JE (2004). X-Ray Absorption Spectroscopy. Comprehensive Coordination Chemistry II. Volume 2. Elsevier Ltd. 159-186. [144] Nilsson B and Brahme A (1983). Relationship Between Kerma and Absorbed Dose in Photon Beams. Acta Radiologica: Oncology. 22: 77-85. [145] Hill R, Keall P, Beckham W, et al. (1998). Photon Buildup in Orthovoltage X-Ray Beams. Australasian Physical & Engineering Sciences in Medicine. 21: 51-56. [146] Yadav G, Yadav R, and Kumar A (2010). Effect of Various Physical Parameters on Surface and Build-Up Dose for 15-MV X-Rays. Journal of Medical Physics. 35: 202-206. [147] Frenot A and Chronakis IS (2003). Polymer Nanofibers Assembled by Electrospinning. Current Opinion in Colloid and Interface Science 8. 8: 64-75. [148] Formhals A (1934). Process and Apparatus for Preparing Artificial Threads. USA Patent No. 1975504 A.   [149] Darrell HR and Iksoo C (1996). Nanometre Diameter Fibres of Polymer Produced by Electrospinning. Nanotechnology. 7: 216-223.  [150] Larrondo L and John Manley R St (1981). Electrostatic Fiber Spinning from Polymer Melts. I. Experimental Observations on Fiber Formation and Properties. Journal of Polymer Science: Polymer Physics Edition. 19: 909-920. [151] Bhardwaj N and Kundu SC (2010). Electrospinning: A Fascinating Fiber Fabrication Technique. Biotechnology Advances. 28: 325-347. [152] Ashammakhi N, Ndreu A, Piras AM, et al. (2007). Biodegradable Nanomats Produced by Electrospinning: Expanding Multifunctionality and Potential for Tissue Engineering. Journal Nanoscience and Nanotechnology. 7: 862-882. [153] Taylor G (1969). Electrically Driven Jets. Proceedings of the Royal Society of London A Mathematical and Physical Sciences. 313: 453-475. [154] Doshi J and Reneker DH (1995). Electrospinning Process and Applications of Electrospun Fibers. Journal of Electrostatics. 35: 151-160. [155] Yarin AL, Koombhongse S, and Reneker DH (2001). Bending Instability in Electrospinning of Nanofibers. Journal of Applied Physics. 89: 3018-3026. [156] Reneker DH, Yarin AL, Fong H, et al. (2000). Bending Instability of Electrically Charged Liquid Jets of Polymer Solutions in Electrospinning. Journal of Applied Physics. 87: 4531-4547. [157] Li W-J, Laurencin CT, Caterson EJ, et al. (2002). Electrospun Nanofibrous Structure: A Novel Scaffold for Tissue Engineering. Journal of Biomedical Materials Research. 60: 613-621. [158] Matthews JA, Wnek GE, Simpson DG, et al. (2002). Electrospinning of Collagen Nanofibers. Biomacromolecules. 3: 232-238. [159] Frantz C, Stewart KM and Weaver VM (2010). The Extracellular Matrix at a Glance. Journal of Cell Science. 123: 4195-4200. [160] Burger C, Hsiao BS and Chu B (2006). Nanofibrous Materials and Their Applications. Annual Review of Materials Research. 36: 333-368. [161] Kumbar SG, James R, Nukavarapu SP, et al. (2008). Electrospun Nanofiber Scaffolds: Engineering Soft Tissues. Biomedical Materials. 3: 034002. [162] Venugopal J, Low S, Choon AT, et al. (2008). Interaction of Cells and Nanofiber Scaffolds in Tissue Engineering. Journal of Biomedical Materials Research, Part B: Applied Biomaterials. 84B: 34-48. [163] Hu X, Liu S, Zhou G, et al. (2014). Electrospinning of Polymeric Nanofibers for Drug Delivery Applications. Journal of Controlled Release. 185: 12-21. [164] Son YJ, Kim WJ, and Yoo HS (2014). Therapeutic Applications of Electrospun Nanofibers for Drug Delivery Systems. Archives of Pharmacal Research. 37: 69-78. [165] Vasita R, and Katti DS (2006). Nanofibers and their Applications in Tissue Engineering. International Journal of Nanomedicine. 1: 15-30. 123  [166] Rezaei A, Nasirpour A and Fathi M (2015). Application of Cellulosic Nanofibers in Food Science Using Electrospinning and Its Potential Risk. Comprehensive Reviews in Food Science and Food Safety. 14: 269-284. [167] Kriegel C, Arrechi A, Kit K, et al. (2008). Fabrication, Functionalization, and Application of Electrospun Biopolymer Nanofibers. Critical Reviews in Food Science and Nutrition. 48: 775-797. [168] Kim TG and Park TG (2006). Surface Functionalized Electrospun Biodegradable Nanofibers for Immobilization of Bioactive Molecules. Biotechnology Progress. 22: 1108-1113. [169] Nagata S, Atkinson GM, Pestov D, et al. (2013). Electrospun Polymer-Fiber Solar Cell. Advances in Materials Science and Engineering. 2013: 975947. [170] Dai Y, Liu W, Formo E, et al. (2011). Ceramic Nanofibers Fabricated by Electrospinning and their Applications in Catalysis, Environmental Science, and Energy Technology. Polymers for Advanced Technologies. 22: 326-338. [171] Thavasi V, Singh G, and Ramakrishna S (2008). Electrospun Nanofibers in Energy and Environmental Applications. Energy & Environmental Science. 1: 205-221. [172] Nurfaizey AH, Tucker N, Stanger J, et al. (2012). Functional Nanofibers in Clothing for Protection Against Chemical and Biological Hazards. In: Wei Q (ed). Functional Nanofibers and their Applications. 1st Edition. Woodhead Publishing. 236-261. [173] Joshi M and Bhattacharyya A (2011). Nanotechnology - A New Route to High-Performance Functional Textiles. Textile Progress. 43: 155-233. [174] Lee S and Obendorf SK (2007). Use of Electrospun Nanofiber Web for Protective Textile Materials as Barriers to Liquid Penetration. Textile Research Journal. 77: 696-702. [175] Zanin MHA, Cerize NNP and de Oliveira AM (2011). Production of Nanofibers by Electrospinning Technology: Overview and Application in Cosmetics. In: Beck R, Guterres S and Pohlmann A (eds). Nanocosmetics and Nanomedicines: New Approaches for Skin Care. Springer. Heidelberg, Germany: 311-332. [176] Taepaiboon P, Rungsardthong U and Supaphol P (2007). Vitamin-Loaded Electrospun Cellulose Acetate Nanofiber Mats as Transdermal and Dermal Therapeutic Agents of Vitamin A Acid and Vitamin E. European Journal of Pharmaceutics and Biopharmaceutics. 67: 387-397. [177] Cipitria A, Skelton A, Dargaville TR, et al. (2011). Design, Fabrication and Characterization of PCL Electrospun Scaffolds - A Review. Journal of Materials Chemistry. 21: 9419-9453. [178] Huang ZM, Zhang YZ, Kotaki M, et al. (2003). A Review on Polymer Nanofibers by Electrospinning and their Applications in Nanocomposites. Composites Science and Technology. 63: 2223-2253. [179] Croisier F, Duwez AS, Jérôme C, et al. (2012). Mechanical Testing of Electrospun PCL Fibers. Acta Biomaterialia. 8: 218-224. [180] Vert M, Li SM, Spenlehauer G, et al. (1992). Bioresorbability and Biocompatibility of Aliphatic Polyesters. Journal of Materials Science: Materials in Medicine. 3: 432-446. [181] Nair LS and Laurencin CT (2007). Biodegradable Polymers as Biomaterials. Progress in Polymer Science. 32: 762-798. [182] Okada M (2002). Chemical Syntheses of Biodegradable Polymers. Progress in Polymer Science. 27: 87-133. [183] Ligia G and Deodato R (2009). Viscoelastic Behaviour of Polymers. In: Physicochemical Behavior and Supramolecular Organization of Polymers. Springer. Dordrecht, Netherlands: 43-162. [184] Patrício T, Domingos M, Gloria A, et al. (2013). Characterisation of PCL and PCL/PLA Scaffolds for Tissue Engineering. Procedia CIRP. 5: 110-114. [185] Labet M and Thielemans W (2009). Synthesis of Polycaprolactone: A Review. Chemical Society Reviews. 38: 3484-3504. 124  [186] Thadavirul N, Pavasant P and Supaphol P (2014). Development of Polycaprolactone Porous Scaffolds by Combining Solvent Casting, Particulate Leaching, and Polymer Leaching Techniques for Bone Tissue Engineering. Journal of Biomedical Materials Research, Part A. 102: 3379-3392. [187] Yang Q, Chen L, Shen X, et al. (2006). Preparation of Polycaprolactone Tissue Engineering Scaffolds by Improved Solvent Casting/Particulate Leaching Method. Journal of Macromolecular Science, Part B. 45: 1171-1181. [188] Lam CXF, Hutmacher DW, Schantz J-T, et al. (2009). Evaluation of Polycaprolactone Scaffold Degradation for 6 Months in Vitro and in Vivo. Journal of Biomedical Materials Research, Part A. 90A: 906-919. [189] Williams JM, Adewunmi A, Schek RM, et al. (2005). Bone Tissue Engineering Using Polycaprolactone Scaffolds Fabricated via Selective Laser Sintering. Biomaterials. 26: 4817-4827. [190] Salerno A and Domingo C (2015). Pore Structure Properties of Scaffolds Constituted by Aggregated Microparticles of PCL and PCL-HA Processed by Phase Separation. Journal of Porous Materials. 22: 425-435. [191] Oh SH, Park SC, Kim HK, et al. (2011). Degradation Behavior of 3D Porous Polydioxanone-b-Polycaprolactone Scaffolds Fabricated Using the Melt-Molding Particulate-Leaching Method. Journal of Biomaterials Science, Polymer Edition. 22: 225-237. [192] Pappenheim CL and Kirkpatrick B (1988). Cardiac Catheterization. AORN Journal. 48: 1130-1137. [193] Dwyer BK, Taylor L, Fuller A, et al. (2005). Percutaneous Transluminal Coronary Angioplasty and Stent Placement in Pregnancy. Obstetrics & Gynecology. 106: 1162-1164. [194] Lee SE, Lin FY, Lu Y, et al. (2017). Rationale and Design of the Coronary Computed Tomographic Angiography for Selective Cardiac Catheterization: Relation to Cardiovascular Outcomes, Cost Effectiveness and Quality of Life (Conserve) Trial. American Heart Journal. 186: 48-55. [195] Consuegra-Sánchez L, Jaulent-Huertas L, Vicente-Gilabert M, et al. (2017). Effect of Part-Time Cardiac Catheterization Facilities in Patients with Acute Myocardial Infarction. International Journal of Cardiology. 236: 85-90. [196] Adachi Y, Sakakura K, Wada H, et al. (2016). Predictors of Prolonged Fluoroscopy Time in Diagnostic Coronary Angiography. Journal of Cardiology. 68: 37-42. [197] Zaman S, Gooley R, McCormick L, et al. (2015). Pre - Transcatheter Aortic Valve Implantation Workup in the Cardiac Catheterisation Laboratory. Heart, Lung and Circulation. 24: 1162-1170. [198] Ailon  J, Mourad  O, Chien  V, et al. (2014). Ultrasound-Guided Insertion of a Radial Arterial Catheter. New England Journal of Medicine. 371: e21. [199] Balls A, LoVecchio F, Kroeger A, et al. (2010). Ultrasound Guidance for Central Venous Catheter Placement: Results from the Central Line Emergency Access Registry Database. The American Journal of Emergency Medicine. 28: 561-567. [200] Serafimidis K, Sakorafas GH, Konstantoudakis G, et al. (2009). Ultrasound-Guided Catheterization of the Internal Jugular Vein in Oncologic Patients; Comparison with the Classical Anatomic Landmark Technique: A Prospective Study. International Journal of Surgery. 7: 526-528. [201] Karakitsos D, Labropoulos N, De Groot E, et al. (2006). Real-Time Ultrasound-Guided Catheterisation of the Internal Jugular Vein: A Prospective Comparison with the Landmark Technique in Critical Care Patients. Critical Care. 10: R162-R162. [202] Ratnayaka K, Faranesh AZ, Hansen MS, et al. (2013). Real-Time MRI-Guided Right Heart Catheterization in Adults Using Passive Catheters. European Heart Journal. 34: 380-389. [203] Muller L, Saeed M, Wilson MW, et al. (2012). Remote Control Catheter Navigation: Options for Guidance under MRI. Journal of Cardiovascular Magnetic Resonance. 14: 33. [204] Tzifa A, Schaeffter T and Razavi R (2012). MR Imaging-Guided Cardiovascular Interventions in Young Children. Magnetic Resonance Imaging Clinics of North America. 20: 117-128. 125  [205] Settecase F, Sussman MS, Wilson MW, et al. (2007). Magnetically-Assisted Remote Control (MARC) Steering of Endovascular Catheters for Interventional MRI: A Model for Deflection and Design Implications. Medical Physics. 34: 3135-3142. [206] Sahagian R (1999). Critical Insight: Marking Devices with Radiopaque Coatings. Medical Device and Diagnostic Industry, Santa Monica, USA. [207] Tartacower R, Jansen L Gilbert J, et al. (1997). Catheter Guidewire with Radiopaque Markers. USA Patent No. 5606981 A. [208] Hammersmark DJ, Taylor KD and Greenfield SR (1995). Radiopaque Tip Maker for Alignment of a Catheter within a Body. USA Patent No. 5429617 A. [209] Gambale RA and Crittenden JF (1990). Catheter Guidewire with Varying Radiopacity. USA Patent No. 4922924 A. [210] ProPlate®. Medical Coating Solutions. http://proplate.com/medical-coating-solutions.html [Accessed on March 13, 2017].  [211] Putnam Plastics. Polymer Marker Bands. http://www.putnamplastics.com/technologies/marker-bands [Accessed on March 13, 2017]. [212] Becker K, Salar-Behzadi S and Zimmer A (2015). Solvent-Free Melting Techniques for the Preparation of Lipid-Based Solid Oral Formulations. Pharmaceutical Research. 32: 1519-1545. [213] Patil AT, Khobragade DS, Chafle SA, et al. (2012). Development and Evaluation of a Hot-Melt Coating Technique for Enteric Coating. Brazilian Journal of Pharmaceutical Sciences. 48: 69-77. [214] Jaimini M, Jain A, Sharma SK, et al. (2014). Solventless Coating for Tablets: An Alternative to Conventional Coating Technique. Indian Journal of Pharmaceutical and Biological Research. 2: 108-116. [215] Achanta AS, Adusumilli PS, James KW, et al. (1997). Development of Hot Melt Coating Methods. Drug Development and Industrial Pharmacy. 23: 441-449. [216] Saatchi K, Patrick BO and Orvig C (2005). Coordination Chemistry of P, O-Bidentate Phosphinophenolates with Ga3+ and In3+. Dalton Transactions. 2268-2274. [217] Kovacs MS, Hein P, Sattarzadeh S, et al. (2001). Complexes of Phosphine-Phenolate Ligands with the [Re=O]3+ and [Re(HNNC5H4N)(NNC5H4N)]2+ Cores. Journal of the Chemical Society, Dalton Transactions. 3015-3024. [218] American Type Culture Collection (ATCC). HEK-293 (ATCC® CRL-1573TM). https://www.atcc.org/Produ cts/All/CRL-1573.aspx [Accessed on October 21, 2016].  [219] Johnston ST, Shah ET, Chopin LK, et al. (2015). Estimating Cell Diffusivity and Cell Proliferation Rate by Interpreting Incucyte ZoomTM Assay Data Using the Fisher-Kolmogorov Model. BMC Systems Biology. 9: 38. [220] Stewart H, Bartlett C, Ross-Thriepland D, et al. (2015). A Novel Method for the Measurement of Hepatitis C Virus Infectious Titres Using the Incucyte Zoom and Its Application to Antiviral Screening. Journal of Virological Methods. 218: 59-65. [221] Mosmann T (1983). Rapid Colorimetric Assay for Cellular Growth and Survival: Application to Proliferation and Cytotoxicity Assays. Journal of Immunological Methods. 65: 55-63. [222] Korzeniewski C and Callewaert DM (1983). An Enzyme-Release Assay for Natural Cytotoxicity. Journal of Immunological Methods. 64: 313-320. [223] Collins G, Federici J, Imura Y, et al. (2012). Charge Generation, Charge Transport, and Residual Charge in the Electrospinning of Polymers: A Review of Issues and Complications. Journal of Applied Physics. 111: 044701. [224] Cho BM, Nam YS, Cheon JY, et al. (2015). Residual Charge and Filtration Efficiency of Polycarbonate Fibrous Membranes Prepared by Electrospinning. Journal of Applied Polymer Science. 132: 41340. [225] Arita Y, Sha Shiratori S and Ikezaki K (2003). A Method for Detection and Visualization of Charge Trapping Sites in Amorphous Parts in Crystalline Polymers. Journal of Electrostatics. 57: 263-271. 126  [226] Zhong Y, Leung V, Yuqin Wan L, et al. (2015). Electrospun Magnetic Nanofibre Mats - A New Bondable Biomaterial Using Remotely Activated Magnetic Heating. Journal of Magnetism and Magnetic Materials. 380: 330-334. [227] Katsogiannis KAG, Vladisavljević GT and Georgiadou S (2015). Porous Electrospun Polycaprolactone (PCL) Fibres by Phase Separation. European Polymer Journal. 69: 284-295. [228] Koski A, Yim K and Shivkumar S (2004). Effect of Molecular Weight on Fibrous PVA Produced by Electrospinning. Materials Letters. 58: 493-497. [229] Wool RP (1993). Polymer Entanglements. Macromolecules. 26: 1564-1569. [230] Sukigara S, Gandhi M, Ayutsede J, et al. (2003). Regeneration of Bombyx Mori Silk by Electrospinning - Part 1: Processing Parameters and Geometric Properties. Polymer. 44: 5721-5727. [231] Spivak AF, Dzenis YA and Reneker DH (2000). Model of Steady State Jet in the Electrospinning Process. Mechanics Research Communications. 27: 37-42. [232] Levenspiel O (1984). Flow of Incompressible Newtonian Fluids in Pipes. In: Engineering Flow and Heat Exchange. Springer. Boston, USA: 17-39. [233] Nisbet DR, Rodda AE, Finkelstein DI, et al. (2009). Surface and Bulk Characterisation of Electrospun Membranes: Problems and Improvements. Colloids and Surfaces B: Biointerfaces. 71: 1-12. [234] Jha BS, Colello RJ, Bowman JR, et al. (2011). Two Pole Air Gap Electrospinning: Fabrication of Highly Aligned, Three-Dimensional Scaffolds for Nerve Reconstruction. Acta Biomaterialia. 7: 203-215. [235] Schnell E, Klinkhammer K, Balzer S, et al. (2007). Guidance of Glial Cell Migration and Axonal Growth on Electrospun Nanofibers of Poly-ε-Caprolactone and a Collagen/Poly-ε-Caprolactone Blend. Biomaterials. 28: 3012-3025. [236] Gerardo-Nava J, Führmann T, Klinkhammer K, et al. (2008). Human Neural Cell Interactions with Orientated Electrospun Nanofibers in Vitro. Nanomedicine. 4: 11-30. [237] Yoshimoto H, Shin YM, Terai H, et al. (2003). A Biodegradable Nanofiber Scaffold by Electrospinning and its Potential for Bone Tissue Engineering. Biomaterials. 24: 2077-2082. [238] Zhu Y, Cao Y, Pan J, et al. (2010). Macro-Alignment of Electrospun Fibers for Vascular Tissue Engineering. Journal of Biomedical Materials Research, Part B: Applied Biomaterials. 92B: 508-516. [239] Pektok E, Nottelet B, Tille JC, et al. (2008). Degradation and Healing Characteristics of Small-Diameter Poly(ε-Caprolactone) Vascular Grafts in the Rat Systemic Arterial Circulation. Circulation. 118: 2563-2570. [240] Chong EJ, Phan TT, Lim IJ, et al. (2007). Evaluation of Electrospun PCL/Gelatin Nanofibrous Scaffold for Wound Healing and Layered Dermal Reconstitution. Acta Biomaterialia. 3: 321-330. [241] Erisken C, Kalyon DM and Wang H (2008). Functionally Graded Electrospun Polycaprolactone and β-Tricalcium Phosphate Nanocomposites for Tissue Engineering Applications. Biomaterials. 29: 4065-4073. [242] Li W-J, Danielson KG, Alexander PG, et al. (2003). Biological Response of Chondrocytes Cultured in Three-Dimensional Nanofibrous Poly(ε-Caprolactone) Scaffolds. Journal of Biomedical Materials Research, Part A. 67A: 1105-1114. [243] Pham QP, Sharma U and Mikos AG (2006). Electrospun Poly(ε-Caprolactone) Microfiber and Multilayer Nanofiber/Microfiber Scaffolds:  Characterization of Scaffolds and Measurement of Cellular Infiltration. Biomacromolecules. 7: 2796-2805. [244] Nottelet B, Pektok E, Mandracchia D, et al. (2009). Factorial Design Optimization and in Vivo Feasibility of Poly(ε-Caprolactone) Micro- and Nanofiber-Based Small Diameter Vascular Grafts. Journal of Biomedical Materials Research, Part A. 89A: 865-875. [245] Yang X, Yang F, Walboomers XF, et al. (2010). The Performance of Dental Pulp Stem Cells on Nanofibrous PCL/Gelatin/nHA Scaffolds. Journal of Biomedical Materials Research, Part A. 93A: 247-257. 127  [246] Li X, Xie J, Yuan X, et al. (2008). Coating Electrospun Poly(ε-Caprolactone) Fibers with Gelatin and Calcium Phosphate and their Use as Biomimetic Scaffolds for Bone Tissue Engineering. Langmuir. 24: 14145-14150. [247] Stolte S, Bui HTT, Steudte S, et al. (2015). Preliminary Toxicity and Ecotoxicity Assessment of Methyltrioxorhenium and its Derivatives. Green Chemistry. 17: 1136-1144. [248] Haley TJ and Cartwright FD (1968). Pharmacology and Toxicology of Potassium Perrhenate and Rhenium Trichloride. Journal of Pharmaceutical Sciences. 57: 321-323. [249] Thakur RA, Florek CA, Kohn J, et al. (2008). Electrospun Nanofibrous Polymeric Scaffold with Targeted Drug Release Profiles for Potential Application as Wound Dressing. International Journal of Pharmaceutics. 364: 87-93. [250] Costoya A, Concheiro A and Alvarez-Lorenzo C (2017). Electrospun Fibers of Cyclodextrins and Poly(Cyclodextrins). Molecules. 22: 230. [251] Coldwell DM, Stokes KR and Yakes WF (1994). Embolotherapy: Agents, Clinical Applications, and Techniques. Radiographics. 14: 623-643. [252] Tan KT and Kachura JR (2006). Embolotherapy for Neuroendocrine Tumor Hepatic Metastases. In: Golzarian J, Sun S, and Sharafuddin MJ (eds). Vascular Embolotherapy: A Comprehensive Approach. Volume 2. Oncology, Trauma, Gene Therapy, Vascular Malformations and Neck. Springer. Heidelberg, Germany: 177-188. [253] Park HS, Kim YI, Kim HY, et al. (2007). Bronchial Artery and Systemic Artery Embolization in the Management of Primary Lung Cancer Patients with Hemoptysis. CardioVascular and Interventional Radiology. 30: 638-643. [254] Oklu R and Kalva SP (2012). Embolotherapy for Renal Tumors. In: Mueller P and Adam A (eds). Interventional Oncology: A Practical Guide for the Interventional Radiologist. Springer. New York, USA: 127-130. [255] Albu S, Grigoriu C, Vasiliu C, et al. (2011). The Role of Uterine Artery Embolization in Cervical Cancer - Single Case Report. Mædica. 6: 137-140. [256] Stewart BW and Kleihues P (2003). World Cancer Report. Lyon, France. [257] Canadian Cancer Statistics. http://www.cancer.ca/en/?region=bc [Accessed on September 23, 2016]. [258] Ulmer SC (2000). Hepatocellular Carcinoma: A Concise Guide to its Status and Management. Postgraduate Medicine. 107: 117-124. [259] Chuang SC, Vecchia CL and Boffetta P (2009). Liver Cancer: Descriptive Epidemiology and Risk Factors Other Than HBV and HCV Infection. Cancer Letters. 286: 9-14. [260] Kumar NB (2012). Assessment of Malnutrition and Nutritional Therapy Approaches. In: Nutritional Management of Cancer Treatment Effects. Volume XVII. Springer. New York, USA: 256. [261] Refaat R and Hassan MS (2014). The Relationship Between the Percentage of Lung Shunting on Tc-99m Macroaggregated Albumin (Tc-99m MAA) Scan and the Grade of Hepatocellular Carcinoma Vascularity. The Egyptian Journal of Radiology and Nuclear Medicine. 45: 333-342. [262] Kennedy A, Nag S, Salem R, et al. (2007). Recommendations for Radioembolization of Hepatic Malignancies Using Yttrium-90 Microsphere Brachytherapy: A Consensus Panel Report from the Radioembolization Brachytherapy Oncology Consortium. International Journal of Radiation Oncology * Biology * Physics. 68: 13-23. [263] Salem R and Lewandowski RJ (2013). Chemoembolization and Radioembolization for Hepatocellular Carcinoma. Clinical Gastroenterology and Hepatology. 11: 604-611. [264] Mojtahedi A, Yang X and Goswami GK (2008). Embolotherapy in the Management of Hepatocellular Carcinoma. Seminars in Interventional Radiology. 25: 234-241. [265] Lee E, Pachter HL and Sarpel U (2012). Hepatic Arterial Embolization for the Treatment of Metastatic Neuroendocrine Tumors. International Journal of Hepatology. 2012: 471203. 128  [266] Wáng YXJ, De Baere T, Idée JM, et al. (2015). Transcatheter Embolization Therapy in Liver Cancer: An Update of Clinical Evidences. Chinese Journal of Cancer Research. 27: 96-121. [267] Ferrer Puchol MD, la Parra C, Esteban E, et al. (2011). Comparison of Doxorubicin-Eluting Bead Transarterial Chemoembolization (Deb-Tace) with Conventional Transarterial Chemoembolization (Tace) for the Treatment of Hepatocellular Carcinoma. Radiología (English Edition). 53: 246-253. [268] Varela M, Real MI, Burrel M, et al. (2007). Chemoembolization of Hepatocellular Carcinoma with Drug Eluting Beads: Efficacy and Doxorubicin Pharmacokinetics. Journal of Hepatology. 46: 474-481. [269] Pawlik TM, Reyes DK, Cosgrove D, et al. (2011). Phase II Trial of Sorafenib Combined with Concurrent Transarterial Chemoembolization with Drug-Eluting Beads for Hepatocellular Carcinoma. Journal of Clinical Oncology. 29: 3960-3967. [270] Kehoe S, Amensag S, Looney M, et al. (2015). "Imageable" Zinc-Silicate Glass Microspheres for Transarterial Embolization: A Renal Artery Embolization Study. Biomedical Glasses. 1: 70-79. [271] Duran R, Sharma K, Dreher MR, et al. (2016). A Novel Inherently Radiopaque Bead for Transarterial Embolization to Treat Liver Cancer - a Pre-Clinical Study. Theranostics. 6: 28-39. [272] Golowa YS, Cynamon J, Reinus JF, et al. (2012). Value of Noncontrast CT Immediately after Transarterial Chemoembolization of Hepatocellular Carcinoma with Drug-Eluting Beads. Journal of Vascular and Interventional Radiology. 23: 1031-1035. [273] Hasan M, Kehoe S and Boyd D (2014). Temporal Analysis of Dissolution by-Products and Genotoxic Potential of Spherical Zinc–Silicate Bioglass: “Imageable Beads” for Transarterial Embolization. Journal of Biomaterials Applications. 29: 566-581. [274] Kilcup N, Tonkopi E, Abraham RJ, et al. (2015). Composition-Property Relationships for Radiopaque Composite Materials: Pre-Loaded Drug-Eluting Beads for Transarterial Chemoembolization. Journal of Biomaterials Applications. 30: 93-103. [275] Forner A, Gilabert M, Bruix J, et al. (2014). Treatment of Intermediate-Stage Hepatocellular Carcinoma. Nature Reviews Clinical Oncology. 11: 525-535. [276] Riaz A, Kulik L, Lewandowski RJ, et al. (2009). Radiologic–Pathologic Correlation of Hepatocellular Carcinoma Treated with Internal Radiation Using Yttrium-90 Microspheres. Hepatology. 49: 1185-1193. [277] Kulik LM, Carr BI, Mulcahy MF, et al. (2008). Safety and Efficacy of 90Y Radiotherapy for Hepatocellular Carcinoma with and without Portal Vein Thrombosis. Hepatology. 47: 71-81. [278] Kennedy AS, McNeillie P, Dezarn WA, et al. (2009). Treatment Parameters and Outcome in 680 Treatments of Internal Radiation with Resin 90Y-Microspheres for Unresectable Hepatic Tumors. International Journal of Radiation Oncology*Biology*Physics. 74: 1494-1500. [279] Salem R, Mazzaferro V and Sangro B (2013). Yttrium-90 Radioembolization for the Treatment of Hepatocellular Carcinoma: Biological Lessons, Current Challenges, and Clinical Perspectives. Hepatology. 58: 2188-2197. [280] Hilgard P, Hamami M, Fouly AE, et al. (2010). Radioembolization with Yttrium-90 Glass Microspheres in Hepatocellular Carcinoma: European Experience on Safety and Long-Term Survival. Hepatology. 52: 1741-1749. [281] Sangro B, Carpanese L, Cianni R, et al. (2011). Survival after Yttrium-90 Resin Microsphere Radioembolization of Hepatocellular Carcinoma across Barcelona Clinic Liver Cancer Stages: A European Evaluation. Hepatology. 54: 868-878. [282] Pasciak AS, Bourgeois AC and Bradley YC (2016). A Microdosimetric Analysis of Absorbed Dose to Tumor as a Function of Number of Microspheres Per Unit Volume in 90Y Radioembolization. Journal of Nuclear Medicine. 57: 1020-1026. [283] MDS Nordion (2013). Therasphere® (Yttrium-90 Glass Microspheres) - Package Insert.  [284] About SIR-Spheres® Microspheres. http://www.sirtex.com/au/ [Accessed on November 10, 2016].   129  [285] McCann JW, Larkin AM, Martino LJ, et al. (2012). Radiation Emission from Patients Treated with Selective Hepatic Radioembolization Using Yttrium-90 Microspheres: Are Contact Restrictions Necessary? Journal of Vascular and Interventional Radiology. 23: 661-667. [286] Sangro B, Iñarrairaegui M and Bilbao JI (2012). Radioembolization for Hepatocellular Carcinoma. Journal of Hepatology. 56: 464-473. [287] Murthy R, Nunez R, Szklaruk J, et al. (2005). Yttrium-90 Microsphere Therapy for Hepatic Malignancy: Devices, Indications, Technical Considerations, and Potential Complications. Radiographics. 25: S41-S55. [288] Maus S, Buchholz H, Brochhausen C, et al. (2011). Microparticles Based on Biocompatible, Biodegradable Polymers. Possible Surrogates for Y-90 Labelled Sir-Spheres? The Journal of Nuclear Medicine Meeting Abstracts. 52: 1594. [289] Caine M, McCafferty MS, McGhee S, et al. (2017). Impact of Yttrium-90 Microsphere Density, Flow Dynamics, and Administration Technique on Spatial Distribution: Analysis Using an in Vitro Model. Journal of Vascular and Interventional Radiology. 28: 260-268. [290] Burrill J, Häfeli UO and Liu DM (2011). Advances in Radioembolization - Embolics and Isotopes. Journal of Nuclear Medicine and Radiation Therapy. 2: 1-6. [291] Shen S, DeNardo GL, Yuan A, et al. (1994). Planar Gamma Camera Imaging and Quantitation of Yttrium-90 Bremsstrahlung. Journal of Nuclear Medicine. 35: 1381-1389. [292] Minarik D, Sjögreen-Gleisner K, Linden O, et al. (2010). 90Y Bremsstrahlung Imaging for Absorbed-Dose Assessment in High-Dose Radioimmunotherapy. Journal of Nuclear Medicine. 51: 1974-1978. [293] Kathy W, Dale LB and Clive B (2011). Quantifying Lung Shunting During Planning for Radio-Embolization. Physics in Medicine and Biology. 56: N145 - N152. [294] Ahmadzadehfar H, Sabet A, Biermann K, et al. (2010). The Significance of 99mTc-MAA SPECT/CT Liver Perfusion Imaging in Treatment Planning for 90Y-Microsphere Selective Internal Radiation Treatment. Journal of Nuclear Medicine. 51: 1206-1212. [295] Uliel L, Royal HD, Darcy MD, et al. (2012). From the Angio Suite to the γ-Camera: Vascular Mapping and 99mTc-MAA Hepatic Perfusion Imaging Before Liver Radioembolization - A Comprehensive Pictorial Review. Journal of Nuclear Medicine. 53: 1736-1747. [296] Kennedy AS, Dezarn WA, McNeillie P, et al. (2013). Dosimetry and Dose Calculation. In: Bilbao JI, Reiser MF (eds). Liver Radioembolization with 90Y Microspheres. 2nd Edition. Springer. Heilderberg, Germany: 188. [297] Lambert B and de Klerk J (2006). Clinical Applications of 188Re-Labeled Radiopharmaceuticals for Radionuclide Therapy. Nuclear Medicine Communications. 27: 223-229. [298] Cremonesi M, Ferrari M, Bodei L, et al. (2006). Dosimetry in Peptide Radionuclide Receptor Therapy: A Review. Journal of Nuclear Medicine. 47: 1467-1475. [299] Zhao CX and Middelberg APJ (2011). Two-Phase Microfluidic Flows. Chemical Engineering Science. 66: 1394-1411. [300] Streets AM and Huang Y (2013). Chip in a Lab: Microfluidics for Next Generation Life Science Research. Biomicrofluidics. 7: 011302. [301] Whitesides GM (2006). The Origins and the Future of Microfluidics. Nature. 442: 368-373. [302] Mark D, Haeberle S, Roth G, et al. (2010). Microfluidic Lab-on-a-Chip Platforms: Requirements, Characteristics and Applications. Chemical Society Reviews. 39: 1153-1182. [303] Fu T, Ma Y, Funfschilling D, et al. (2010). Squeezing-to-Dripping Transition for Bubble Formation in a Microfluidic T-Junction. Chemical Engineering Science. 65: 3739-3748. [304] Tran VT, Benoît JP and Venier-Julienne MC (2011). Why and How to Prepare Biodegradable, Monodispersed, Polymeric Microparticles in the Field of Pharmacy? International Journal of Pharmaceutics. 407: 1-11. 130  [305] Bokharaei M, Schneider T, Dutz S, et al. (2016). Production of Monodispersed Magnetic Polymeric Microspheres in a Microfluidic Chip and 3D Simulation. Microfluidics and Nanofluidics. 20: 6. [306] Häfeli UO, Saatchi K, Elischer P, et al. (2010). Lung Perfusion Imaging with Monosized Biodegradable Microspheres. Biomacromolecules. 11: 561-567. [307] Bokharaei M, Saatchi K and Häfeli UO (2017). A Single Microfluidic Chip with Dual Surface Properties for Protein Drug Delivery. International Journal of Pharmaceutics. 521: 84-91. [308] Mead-Hunter R, King AJC and Mullins BJ (2012). Plateau Rayleigh Instability Simulation. Langmuir. 28: 6731-6735. [309] Tan YC, Cristini V and Lee AP (2006). Monodispersed Microfluidic Droplet Generation by Shear Focusing Microfluidic Device. Sensors and Actuators B: Chemical. 114: 350-356. [310] Zhang X (1999). Dynamics of Drop Formation in Viscous Flows. Chemical Engineering Science. 54: 1759-1774. [311] Hackley VA (1999). Nomenclature for Ceramic Particle Dispersion Systems. National Institute of Standards and Technology, Gaithersburg, MD  [312] De La Vega JC, Elischer P, Schneider T, et al. (2013). Uniform Polymer Microspheres: Monodispersity Criteria, Methods of Formation and Applications. Nanomedicine. 8: 265-285. [313] Izumida Y, Sugiura S, Oda T, et al. (2005). Production of Quasi-Monodisperse Emulsions with Large Droplets Using a Micromachined Device. Journal of the American Oil Chemists' Society. 82: 73-78. [314] Saatchi K and Häfeli UO (2009). Radiolabeling of Biodegradable Polymeric Microspheres with [99mTc(CO)3]+ and in Vivo Biodistribution Evaluation Using MicroSPECT/CT Imaging. Bioconjugate Chemistry. 20: 1209-1217. [315] American Type Culture Collection (ATCC). N1-S1 (ATCC® CRL-1604TM). https://www.atcc.org/Products /All/CRL-1604.aspx#generalinformation [Accessed on November 18, 2015].  [316] Huang M and Liu G (1998). The Study of Innate Drug Resistance of Human Hepatocellular Carcinoma Bel7402 Cell Line. Cancer Letters. 135: 97-105. [317] Bartkowski R, Berger MR, Aguiar JLA, et al. (1986). Experiments on the Efficacy and Toxicity of Locoregional Chemotherapy of Liver Tumors with 5-Fluoro-2'-Deoxyuridine (FUDR) and 5-Fluorouracil (5-FU) in an Animal Model. Journal of Cancer Research and Clinical Oncology. 111: 42-46. [318] Corning® Cell Culture Surfaces. www.corning.com/lifesciences [319] Jäger W, Moskalev I, Janssen C, et al. (2013). Ultrasound-Guided Intramural Inoculation of Orthotopic Bladder Cancer Xenografts: A Novel High-Precision Approach. PLoS ONE. 8: e59536. [320] Mullen P (2004). The Use of Matrigel to Facilitate the Establishment of Human Cancer Cell Lines as Xenografts. In: Langdon SP (ed). Cancer Cell Culture: Methods and Protocols. Humana Press. Totowa, USA: 287-292. [321] Fridman R, Benton G, Aranoutova I, et al. (2012). Increased Initiation and Growth of Tumor Cell Lines, Cancer Stem Cells and Biopsy Material in Mice Using Basement Membrane Matrix Protein (Cultrex or Matrigel) Co-Injection. Nature Protocols. 7: 1138-1144. [322] Vauthier C, Schmidt C and Couvreur P (1999). Measurement of the Density of Polymeric Nanoparticulate Drug Carriers by Isopycnic Centrifugation. Journal of Nanoparticle Research. 1: 411-418. [323] Ogawa K (1994). Simulation Study of Triple-Energy-Window Scatter Correction in Combined TI-201, Tc-99m SPECT. Annals of Nuclear Medicine. 8: 277-281. [324] Braat AJAT, Smits MLJ, Braat MNGJA, et al. (2015). 90Y Hepatic Radioembolization: An Update on Current Practice and Recent Developments. Journal of Nuclear Medicine. 56: 1079-1087. [325] Kuo JC, Tazbirkova A, Allen R, et al. (2014). Serious Hepatic Complications of Selective Internal Radiation Therapy with Yttrium-90 Microsphere Radioembolization for Unresectable Liver Tumors. Asia-Pacific Journal of Clinical Oncology. 10: 266-272. 131  [326] Hamoui N and Ryu RK (2011). Hepatic Radioembolization Complicated by Fulminant Hepatic Failure. Seminars in Interventional Radiology. 28: 246-251. [327] Riaz A, Lewandowski RJ, Kulik LM, et al. (2009). Complications Following Radioembolization with Yttrium-90 Microspheres: A Comprehensive Literature Review. Journal of Vascular and Interventional Radiology. 20: 1121-1130. [328] Walrand S, Hesse M, Chiesa C, et al. (2014). The Low Hepatic Toxicity per Gray of 90Y Glass Microspheres is Linked to their Transport in the Arterial Tree Favoring a Nonuniform Trapping as Observed in Posttherapy PET Imaging. Journal of Nuclear Medicine. 55: 135-140. [329] Kennedy AS, Kleinstreuer C, Basciano CA, et al. (2010). Computer Modeling of Yttrium-90 Microsphere Transport in the Hepatic Arterial Tree to Improve Clinical Outcomes. International Journal of Radiation Oncology*Biology*Physics. 76: 631-637. [330] Vitello DJ, Ripper RM, Fettiplace MR, et al. (2015). Blood Density is Nearly Equal to Water Density: A Validation Study of the Gravimetric Method of Measuring Intraoperative Blood Loss. Journal of Veterinary Medicine. 2015: 152730. [331] Kokabi N, Galt JR, Xing M, et al. (2014). A Simple Method for Estimating Dose Delivered to Hepatocellular Carcinoma after Yttrium-90 Glass-Based Radioembolization Therapy: Preliminary Results of a Proof of Concept Study. Journal of Vascular and Interventional Radiology. 25: 277-287. [332] Lam MGEH, Banerjee S, Louie JD, et al. (2013). Root Cause Analysis of Gastroduodenal Ulceration after Yttrium-90 Radioembolization. CardioVascular and Interventional Radiology. 36: 1536-1547. [333] Lord R (1999). Osmosis, Osmometry, and Osmoregulation. Postgraduate Medical Journal. 75: 67-73. [334] Anayama T, Nakajima T, Dunne M, et al. (2013). A Novel Minimally Invasive Technique to Create a Rabbit VX2 Lung Tumor Model for Nano-Sized Image Contrast and Interventional Studies. PLoS ONE. 8: e67355. [335] Wang D, Bangash AK, Rhee TK, et al. (2007). Liver Tumors: Monitoring Embolization in Rabbits with VX2 Tumors - Transcatheter Intraarterial First-Pass Perfusion MR Imaging. Radiology. 245: 130-139. [336] Hong K, Khwaja A, Liapi E, et al. (2006). New Intra-Arterial Drug Delivery System for the Treatment of Liver Cancer: Preclinical Assessment in a Rabbit Model of Liver Cancer. Clinical Cancer Research. 12: 2563-2567. [337] Hough A Jr., Seyberth H, Oates J, et al. (1977). Changes in Bone and Bone Marrow in VX2-Bearing Rabbits: A Comparison of Local and Distant Effects. American Journal of Pathology. 87: 537-552. [338] Chen JH, Lin YC, Huang YS, et al. (2004). Induction of VX2 Carcinoma in Rabbit Liver: Comparison of Two Inoculation Methods. Laboratory Animals. 38: 79-84. [339] Nass N, Streit S, Wybranski C, et al. (2017). Validation of VX2 as a Hepatocellular Carcinoma Model: Comparison of the Molecular Reaction of VX2 and HepG2 Tumor Cells to Sorafenib in Vitro. Anticancer Research. 37: 87-93. [340] Alver B, Jauert PA, Brosnan L, et al. (2013). A Whole Genome Screen for Minisatellite Stability Genes in Stationary-Phase Yeast Cells. G3: Genes|Genomes|Genetics. 3: 741-756. [341] Martinez MJ, Roy S, Archuletta AB, et al. (2004). Genomic Analysis of Stationary-Phase and Exit in Saccharomyces Cerevisiae: Gene Expression and Identification of Novel Essential Genes. Molecular Biology of the Cell. 15: 5295-5305. [342] Ashrafi K, Sinclair D, Gordon JI, et al. (1999). Passage Through Stationary Phase Advances Replicative Aging in Saccharomyces Cerevisiae. Proceedings of the National Academy of Sciences of the United States of America. 96: 9100-9105. [343] Takeuchi M, Takeuchi K, Ozawa Y, et al. (2009). Aneuploidy in Immortalized Human Mesenchymal Stem Cells with Non-Random Loss of Chromosome 13 in Culture. In Vitro Cellular & Developmental Biology - Animal. 45: 290-299. 132  [344] Takeuchi M, Higashino A, Takeuchi K, et al. (2015). Correction: Transcriptional Dynamics of Immortalized Human Mesenchymal Stem Cells During Transformation. PLoS ONE. 10: e0131383. [345] Leung WT, Lau WY, Ho SKW, et al. (1994). Measuring Lung Shunting in Hepatocellular Carcinoma with Intrahepatic-Arterial Technetium-99m Macroaggregated Albumin. Journal of Nuclear Medicine. 35: 70-73. [346] Wright CL, Werner JD, Tran JM, et al. (2012). Radiation Pneumonitis Following Yttrium-90 Radioembolization: Case Report and Literature Review. Journal of Vascular and Interventional Radiology. 23: 669-674. [347] Leung WT, Lau WY, Ho SKW, et al. (1995). Radiation Pneumonitis after Selective Internal Radiation Treatment with Intraarterial 90Y-Microspheres for Inoperable Hepatic Tumors. International Journal of Radiation Oncology*Biology*Physics. 33: 919-924. [348] Chiesa C, Mira M, Maccauro M, et al. (2012). A Dosimetric Treatment Planning Strategy in Radioembolization of Hepatocarcinoma with 90Y Glass Microspheres. The Quarterly Journal of Nuclear Medicine and Molecular Imaging. 56: 503-508. [349] Peker A, Çiçek O, Soydal Ç, et al. (2015). Radioembolization with Yttrium-90 Resin Microspheres for Neuroendocrine Tumor Liver Metastases. Diagnostic and Interventional Radiology. 21: 54-59. [350] Riaz A, Awais R and Salem R (2014). Side Effects of Yttrium-90 Radioembolization. Frontiers in Oncology. 4: 198. [351] Moiseenko V, Banáth JP, Duzenli C, et al. (2008). Effect of Prolonging Radiation Delivery Time on Retention og GammaH2AX. Radiation Oncology. 3: 18. [352] Moiseenko V, Duzenli C and Durand RE. (2007). In Vitro Study of Cell Survival Following Dynamic MLC Intensity-Modulated Radiation Therapy Dose Delivery. Medical Physics. 34: 1514-1520. [353] Bergman AM, Bush K, Milette M-P, et al. (2006). Direct Aperture Optimization for IMRT Using Monte Carlo Generated Beamlets. Medical Physics. 33: 3666-3679.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0354674/manifest

Comment

Related Items