UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development and evaluation of an interface pressure transducer for biomedical applications Paris-Seeley, Nancy J. 1996

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

Item Metadata

Download

Media
831-ubc_1996-0010.pdf [ 13MB ]
Metadata
JSON: 831-1.0087030.json
JSON-LD: 831-1.0087030-ld.json
RDF/XML (Pretty): 831-1.0087030-rdf.xml
RDF/JSON: 831-1.0087030-rdf.json
Turtle: 831-1.0087030-turtle.txt
N-Triples: 831-1.0087030-rdf-ntriples.txt
Original Record: 831-1.0087030-source.json
Full Text
831-1.0087030-fulltext.txt
Citation
831-1.0087030.ris

Full Text

DEVELOPMENT AND EVALUATION OF AN INTERFACE PRESSURE TRANSDUCER FOR BIOMEDICAL APPLICATIONS By Nancy J. Paris-Seeley, PEng BASc, University of British Columbia Vancouver, Canada, 1990. A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR T H E D E G R E E OF MASTERS OF APPLIED SCIENCE in T H E F A C U L T Y OF GRADUATE STUDIES Department of Mechanical Engineering We accept this thesis as conforming to the reguired^standard T H E UNIVERSITY OF BRITISH COLUMBIA November, 1995 ®Nancy J. Paris-Seeley, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of ^e.cWm(fli E W H Etfl'1*1) The University of British Columbia Vancouver, Canada Date DE-6 (2/88) 11 A B S T R A C T The measurement of the interface pressure between a biomedical device and part of the human body is useful to aid in the design or improve the performance and safety of such devices. Therefore, a need exists for a transducer to measure interface pressure in these applications. The development and evaluation of an interface pressure transducer was the main goal of this research. Surgical retraction, surgical tourniquets and mammography were selected as demonstration applications for the developed transducer. These target applications were selected because they represented a wide spectrum of device and tissue characteristics and properties, and were in common use. A review of the available clinical, commercial and engineering literature identified a wide range of transducers and transducer technologies used for interface pressure measurement. The transducers included pneumatic/hydraulic, fibre-optic, strain based, capacitive and micro-machined technologies. No standard method of measuring interface pressure was described and, in many cases, investigators cautioned against comparing-interface pressure measurements obtained using different measurement systems. From this review and an examination of the biomedical applications mentioned, the design criteria and optimal design specifications for an interface pressure transducer were defined. To gain a better understanding of the mechanical response of the interface between a device, transducer, and tissue to an applied loading, a preliminary finite element model was developed and studied. The model demonstrated the potential for shear stresses to develop between the transducer and interface materials. Furthermore a calibration system which simulated interface conditions was developed to evaluate both existing and developed transducers for use as interface pressure transducers. This evaluation demonstrated the lack of a Ill transducer whose output was independent of the compliance of the interface materials. As well, an essential characteristic was identified for an effective interface pressure transducer that could be used in several applications where the interface material compliance was different. Based on the knowledge gained from the finite element analysis and existing transducer evaluation results, a novel interface pressure transducer was developed and evaluated both in the calibration system and via demonstration applications of surgical retraction and tourniquets. Under laboratory conditions in the calibration system, the transducer met many of the desired design specifications. The transducer was tested in the lab under both pneumatic and non-pneumatic tourniquet cuffs. The transducer worked well under the pneumatic cuff but required ] further development for use under the non-pneumatic cuff. The transducer was also integrated into a surgical retractor and evaluated in five clinical trials. It met many of the desired specifications for this application. iv T A B L E O F C O N T E N T S ABSTRACT ii LIST OF FIGURES vii LIST OF TABLES ix ACKNOWLEDGMENTS x 1 INTRODUCTION 1 1.1 Motivation 1 1.2 Thesis objectives 5 1.3 Thesis overview 6 2 B A C K G R O U N D AND REVIEW OF PREVIOUS RESEARCH 8 2.1 Chapter overview 8 2.2 Target applications 9 2.2.1 Surgical retractors 10 2.2.2 Tourniquets 15 2.2.3 Mammography 18 2.3 Review of literature of other biomedical interface pressure measurement applications 22 2.3.1 Cushion and mattress applications 23 2.3.2 Prosthetic sockets 25 2.3.3 Foot pressure 28 2.3.4 Pressure garments for treatment of burn injuries 29 2.3.5 Summary of review 30 2.4 Design criteria and specifications for interface pressure transducers 31 2.4.1 Generic criteria for transducers 31 2.4.2 Specific criteria for an interface pressure transducer 33 2.4.3 Functional specifications for an optimal biomedical interface pressure transducer 3 5 2.5 Summary 37 3 FINITE E L E M E N T M O D E L OF MEASUREMENT ENVIRONMENT AND TRANSDUCER 39 3.1 Chapter overview 39 3.2 Brief overview of the finite element method 41 3.3 Development of the transducer and interface material finite element mode!43 3.3.1 Model assumptions 44 3.3.2 Model verification 51 3.3.3 Applied loading, finite element solution, and discussion 51 3.3.3.1 Flush transducer model 52 3.3.3.2 Protruding transducer model 5 5 3.4 Summary 58 4 D E V E L O P M E N T OF A CALIBRATION SYSTEM AND E V A L U A T I O N OF INTERFACE PRES SURE TRANSDUCERS 62 4.1 Chapter overview 62 4.2 Calibration system 63 4.3 Evaluation of existing interface transducers 66 4.3.1 Experimental design 67 4.3.2 Transducers 68 4.3.2.1 Transducers based on force sensitive resistors 68 4.3.2.2 Capacitive transducer 71 4.3.2.3 Intracranial pressure transducer with strain gauged diaphragm 72 4.3.2.4 Transducers having constant pneumatic flow 72 4.3.2.5 Transducers employing fluid-filled pads with remote fluid pressure measurement 74 4.3.3 Test procedures, results and discussion 76 4.3.3.1 Transducers based on force sensitive resistors 76 4.3.3.2 Capacitive transducer 77 4.3.3.3 Intracranial pressure transducer with strain gauged diaphragm 80 4.3.3.4 Transducers having constant pneumatic flow 81 4.3.3.5 Transducers employing fluid-filled pads with remote fluid pressure measurement 83 4.4 Conclusions 87 4.5 Summary 88 5 D E V E L O P M E N T OF A N O V E L INTERFACE PRES SURE TRANSDUCER 89 5.1 Chapter overview 89 5.2 Underlying principle of transducer design 90 5.3 Transducer operating principles, components and instrumentation 91 5.3.1 Transducer operation 91 5.3.2 Transducer components 92 5.3.3 Instrumentation 93 vi 5.4 Transducer prototypes 94 5.4.1 Proof-of-concept prototype 94 5.4.2 Second generation prototype transducer 97 5.4.3 Results of second generation transducer testing 100 5.4.4 Third generation prototype transducer 102 5.4.5 Results of third generation transducer testing 104 5.4.6 Fourth generation prototype transducer 107 5.4.7 Results of fourth generation transducer testing 109 5.4.8 Comparison to optimal specifications 111 5.5 Summary 113 6 E V A L U A T I O N OF T H E DEVELOPED TRANSDUCER IN DEMONSTRATION APPLICATIONS 115 6.1 Chapter overview 115 6.2 Clinical evaluation of the novel transducer on a surgical retractor blade 117 6.2.1 Objectives of surgical retractor trials 118 6.2.2 Materials and methods 121 6.2.3 Results and discussion 121 6.2.4 Conclusions 124 6.3 Laboratory evaluation of the novel transducer under pneumatic and non-pneumatic tourniquet cuffs 126 6.3.1 Transducer performance under pneumatic cuff 126 6.3.2 Transducer performance under non-pneumatic cuff 129 6.4 Summary 131 7 CONCLUSIONS AND RECOMMENDATIONS 133 7.1 General conclusions 133 7.2 Contributions of the research 138 7.3 Recommendations for further work 139 REFERENCES 142 APPENDIX A: Finite element model verification and program code 148 LIST OF FIGURES Figure 2 1 Hand held, self-retaining, and automated retractors 11 Figure 2 2 Tourniquet system 15 Figure 2 3 Mammography procedure and corresponding mammogram 20 Figure 3 1 Cross-section of the physical model of flush transducer, device and tissue 44 Figure 3 2 Cross-section of the physical model of protruding transducer, device and tissue 45 Figure 3 3 Finite element model of the flush transducer and interface materials 47 Figure 3 4 Finite element model of the protruding transducer and interface materials 48 Figure 3 5 Applied loading 52 Figure 3 6 Displacement of the flush transducer model due to applied loading 53 Figure 3 7 Stresses in the x-direction due to applied loading on the flush transducer 54 Figure 3 8 Stresses in the y-direction due to applied loading on the flush transducer 56 Figure 3 9 Displacement of the protruding transducer model due to applied loading 57 Figure 3 10 Stresses in the x-direction due to applied loading on the protruding transducer 59 Figure 3 11 Stresses in the y-direction due to applied loading on the protruding transducer 60 Figure 4 1 Schematic of the interface measurement environment 64 Figure 4 2 Calibration system designed by Sachs 65 Figure 4 3 Components of the developed calibration system 66 Figure 4 4 Transducer orientations and interface material lay-ups used in the calibration system 69 Figure 4 5 Uniforce™ transducer schematic 70 Figure 4 6 FSR™ transducer schematic - unfolded view 71 Figure 4 7 Constant pneumatic flow transducer schematic 73 Figure 4 8 Steritek® transducer schematic 74 Figure 4 9 Oil-filled transducer system schematic 75 Figure 4 10 Pneumatic electric switch transducer schematic 75 Figure 4 11 Input/output characteristics of the Uniforce™ transducer fortest condition 1 77 Figure 4 12 Input/output characteristics of the Nam Tai transducer for test condition 1 78 Figure 4 13 Input/output characteristics of the Nam Tai transducer for test condition 2 78 Figure 4 14 Input/output characteristics of the Nam Tai transducer for test conditions 3 and 4 79 Figure 4 15 Input/output characteristics of the Mikro-tip transducer for test condition 1 80 Figure 4 16 Input/output characteristics of the Mikro-tip transducer for test conditions 2, 3, and 4 82 Vlll Figure 4.17 Input/output characteristics of the constant pneumatic flow transducer prototype for test conditions 1, 3, and 4 83 Figure 4.18 Input/output characteristics of the Steritek® transducer 84 Figure 4.19 Input/output characteristics of the oil-filled pad for test conditions 1, 2, and 3 85 Figure 4.20 Input/output characteristics of the smaller oil-filled pad under test condition 1 85 Figure 4.21 Input/output characteristics of the pneumatic electric switch transducer for test conditions 1 and 2 86 Figure 5.1 Load transfer through material to transducer 90 Figure 5.2 Wheatstone bridge circuit 93 Figure 5.3 Measured relationship between applied pressure and novel transducer output without equalizing pressure for test condition 1 96 Figure 5.4 Measured relationship between applied pressure and novel transducer output with equalizing pressure for test condition 1 96 Figure 5.5 Shematic of transducer and shim 99 Figure 5.6 Measured relationship between applied pressure and the second generation output 100 Figure 5.7 Measured relationship between applied pressure and third generation prototype output 105 Figure 5.8 Measured relationship between applied pressure and third generation prototypes output using test condition 1 105 Figure 5.9 Measured relationship between applied pressure and the third generation prototype output 106 Figure 5.10 Schematic of the fourth generation transducer design 108 Figure 5.11 Measured relationship between applied pressure and the fourth generation prototypes's outputs 110 Figure 5.12 Measured relationship between applied pressure and the fourth generation steel transducer prototype with liquid gel and aluminum shim 110 Figure 6.1 Transducer integrated into shim and bonded to the retractor blade 118 Figure 6.2 Retractor blades retracting tissue 119 Figure 6.3 Tissue sample zones 121 Figure 6.4 First displacement-controlled trial retraction pressures 122 Figure 6.5 Second and third displacement-controlled trial retraction pressures 123 Figure 6.6 Retractor blade displacements for pressure-controlled trials 123 Figure 6.7 Measured relationship between the cuff inflation pressure and the transducer output 128 Figure 6.8 Cross-sectional view of the placement of the transducer and shim under the edge of the cuff 128 Figure 6.9 Effect on applied pressure distribution due to a flat spot 130 Figure 6.10 Pressure distribution with curved section restored 131 ix LIST O F T A B L E S Table 3.1 Mechanical properties of the materials 46 Table 4.1 Comparison of transducers tested 87 Table 5.1 Comparison of initial and second generation prototype 101 Table 5.2 Comparison of second and third generation prototypes 107 Table 5.3 Comparison of third and fourth generation prototypes 111 Table 6.1 Interface pressure measurements under an Esmarch bandage 129 ACKNOWLEDGMENTS I would like to thank my joint supervisors, Jim McEwen and Doug Romilly, for their combination of skills and talents which have guided and assisted me throughout my thesis. Dr. McEwen helped me to see beyond the inevitable day-to-day difficulties to the potential benefits of my work. Dr. Romilly provided me with the necessary mechanical engineering expertise to pull me out of the bog I often seemed to find myself in. I also with to thank Dr. McEwen for providing financial support for this research through the NRC Industrial Research Assistance Program and Dr. Romilly for his financial contribution to this research. The people who make up Western Clinical Engineering gave me much assistance, both technical and otherwise. Ken Glinz was very helpful with my initial transducer evaluation and acted as an interested, often sympathetic, sounding board for my research. Mike Jameson was also helpful with the instrumentation necessary to operate many of the transducers. Alexei Marko provided me with an opportunity to see someone complete and successfully defend their thesis. As well, Martine Janiki reviewed Chapter 4 and provided me with useful advise on writing my thesis. I would also like to thank Anton Schreinders and David Camp, Department of Mechanical Engineering machine shop, for their rnachining of my calibration system and transducer bodies. The quick turn around on my work requests helped to speed my research along. The clinical evaluation of the transducer in a surgical retraction application would not have been possible without the assistance of Dr. Karim Qayumi, Department of Surgery, Vancouver Hospital and Health Sciences Centre (VHHSC). I wish to thank Dr. Qayumi for allowing me to participate in an on-going trial and conducting the surgery for the trial. I also wish to thank Dr. D.A. Owen, Head, Anatomical Pathology, VHHSC, for his assistance in analyzing the tissue samples. I've been fortunate to have a mentor and friend, Judy Findlay, who has been through this process and who has provided me with much support throughout my thesis. I have found that discussing my difficulties with her have invariably ended up in bursts and laughter, greatly reducing my stress level. Lee Giuricich also deserves special thanks for sharing my ups and downs. I am very grateful to my parents and family for their continual support and encouragement. Finally I'd like to thank my biggest supporter, my husband Jon, who has provided me with support, encouragement, and enthusiasm for my research. 1 1 INTRODUCTION 1.1 Motivation Many biomedical procedures involve the application of pressure from a biomedical device to body tissue. Examples include: surgical retractors which apply pressure to move tissue and open a surgical field; tourniquets which apply pressure on the circumference of a limb to occlude blood flow; and mammography where a plate applies pressure to compress breast tissue to increase the probability of detecting a cancerous growth. As well, several rehabilitation areas involve the application o f pressure from body tissue to a support device. Examples include, prosthetic leg sockets where the stump of an amputated leg applies pressure to the socket; wheelchairs where the body applies pressure to the seat; and beds where the body applies pressure to the mattress areas in contact. In all cases, a pressure distribution exists at the interface between the device and tissue. The magnitude and distribution of this interface pressure is of importance in the design, performance, safety, and evaluation of these devices. Interface pressure measurement between body tissue and support structures (such as beds) can aid in the design of a device that can reduce, i f not eliminate, the problem of pressure sores. In addition to device design, interface pressure measurement can also provide information to the user during operation, which could potentially be used for device control. For example, an Chapter 1: Introduction 2 interface pressure transducer is essential for the successful development of an automated retractor. Currently, interface pressure between a retractor and tissue is not measured during surgery. As retraction is necessary in almost all surgical procedures, there is a strong motivation to understand the relationship between retraction pressures and tissue injuries. This relationship could then be utilized to design an automated surgical retractor to reduce, if not eliminate, injuries caused by retractors. Furthermore, a human-held surgical retractor could also benefit from the integration of an interface pressure transducer, particularly in surgeries involving the retraction of brain or liver tissue where tissue damage could lead to permanent debilitating injuries. As previously mentioned, a tourniquet cuff applies pressure around the circumference of the limb to stop blood flow past the cuff. Currently the only indication of interface pressure between the cuff and underlying tissue is the pneumatic pressure in the cuff. However, as the interface pressure has been shown to vary across the width of the cuff, a need exists for an interface pressure transducer which can measure this varying interface pressure. The use of interface pressure transducers could aid in the design of tourniquets cuffs by helping to achieve the most desirable pressure distribution. Such a transducer, if integrated into a tourniquet cuff system, could also be useful in the clinical application of the cuff by helping to ensure the best fit between the cuff and the limb. Furthermore, the pressure measurement data could be integrated into algorithms which could be developed to warn clinicians of high pressures and long application times, thus mitigating the risk of injury due to tourniquet cuffs. Chapter 1: Introduction 3 Non-pneumatic tourniquets have been used to occlude blood flow.in many countries, particularly where the cost of a pneumatic tourniquet is prohibitive. Currently, the interface pressure between the non-pneumatic cuff and the underlying tissue is not measured. The lack of a pressure indicator could potentially result in interface pressures well above or below those required for safe occlusion of blood flow. The integration of a low-cost interface pressure transducer into a simple non-pneumatic tourniquet would improve the efficacy and safety of procedures requiring blood flow occlusion while keeping cost to a minimum. Another application for an interface pressure transducer is in mammography where pressure is required to compress the breast tissue during imaging. Compression is necessary to improve the quality of the image and minimize the amount of radiation required. However, concern has been raised regarding the potential for mammographic breast compression to disrupt an existing tumour and thus increase the spread of cancerous cells. A recent study [1] demonstrated that the pressure indicated on the mammography machine did not correlate with the pressure measured at the interface between the plate of the mammography machine and the breast tissue, demonstrating an inability to provide a desired level of compression. Clearly a need exists for an interface pressure transducer which can accurately measure and control the pressure being applied during a mammogram. An essential characteristic of interface pressure measurements is that the presence of the transducer must not significantly distort the existing pressure distribution. Furthermore, the transducer must be able to conform to the shape of the device/tissue interface. In some cases, Chapter 1: Introduction 4 such as with tourniquets, the interface between the device and tissue may have a radius of curvature which ranges from 20 mm to 200 mm. Therefore the shape, thickness, and flexibility of the transducer are important characteristics. A general purpose interface pressure transducer should produce reliable and repeatable output in several target applications. The compliance and curvature conditions in surgical retraction, mammography, and tourniquet applications vary widely. In mammography, the plate is rigid and flat, and the tissue is very compliant. In surgical retraction, retractors vary in curvature and are rigid, while the tissue may vary in compliance. Lastly, in tourniquets, the cuff varies in curvature and is compliant, while the tissue may vary in compliance from very compliant (i.e. fat) to very stiff (i.e. muscle). Therefore varying compliance and curvature conditions must be addressed when defining transducer specifications. A wide variety of interface measurement transducers have been used in the biomedical and rehabilitation application areas previously mentioned. Results from different types of transducers have been difficult to compare quantitatively, even within one application area. No standards currently exist that define the methods to be used in interface pressure measurement. A need exists for the development of an interface pressure transducer that can be used in several target applications with varying tissue and device compliance wherein the results can be reliably replicated and compared. Chapter 1: Introduction 5 1.2 Thesis objectives The main goal of the research described in this thesis has been to develop and evaluate a transducer that can accurately measure the pressure at the interface of a biomedical device and tissue, and to demonstrate its use in target applications. The following specific objectives were pursued to accomplish this goal: 1) To conduct a comprehensive review of the clinical, engineering, and commercial literature to identify technologies used in interface pressure measurement, and the limitations of available transducers based on those technologies; 2 ) To define optimal specifications for a generic biomedical interface pressure transducer for use in the measurement of pressure at the interface between medical devices and living tissues, organs, and limbs of varying compliance; 3 ) To model the measurement environment and transducer to gain a better understanding of the mechanical response of the system comprised of the transducer and the measurement environment; 4) To design and manufacture a system for transducer calibration and assessment which closely simulates interface material environments of varying compliance, and develop a protocol for evaluating transducer technologies which might satisfy the above specifications; Chapter 1: Introduction 6 5) To identify potential technologies, construct or select appropriate transducers, and evaluate their capabilities in the calibration system with a variety of interface materials of varying compliance; 6) To select or develop a suitable transducer technology and construct a prototype device which best meets the optimal specifications for a generic biomedical interface pressure transducer; 7) To critically assess the prototype transducer in the calibration system under interface materials of varying compliance and compare its performance with the optimal design specifications; and 8) To further evaluate the transducer under clinical conditions in demonstration applications basing the evaluation criteria on the optimal design specifications. 1.3 Thesis overview To assist in achieving these objectives, an initial literature review was performed to provide background for the research work. A summary and evaluation of the available literature review is provided (Chapter 2) and includes a discussion of biomedical applications where Chapter 1: Introduction 7 interface pressure measurements are used, as well as a review of transducers currently used for interface pressure measurements. With this review complete, the design criteria and specifications for an optimal interface pressure transducer were then defined. To better understand the sensitivity of the interface to both material properties and transducer geometry, a simple finite element model of a generic interface was developed (Chapter 3). Based on an improved understanding of the measurement interface, a calibration system was designed and manufactured to evaluate transducer technologies (Chapter 4). Utilizing the literature review and design specifications, existing transducer technologies were identified for potential use in interface pressure transducers. Based on these technologies, transducers were obtained or constructed and evaluated in the calibration system under laboratory conditions (Chapter 4). The evaluation of the existing transducers led to the discovery of an essential characteristic of an interface pressure transducer to be effective in applications with different interface materials. A novel interface pressure transducer was then developed, incorporating this essential characteristic into its design, and evaluated in the calibration system under laboratory conditions (Chapter 5). The integration of the transducer into application devices and subsequent clinical evaluation was then conducted to demonstrate the transducer's potential applicability in the target applications (Chapter 6). This research concludes with general conclusions, contributions of the research, and proposed recommendations for future work (Chapter 7). 8 2 B A C K G R O U N D A N D R E V I E W O F PREVIOUS R E S E A R C H 2.1 Chapter Overview The information in this chapter expands on the motivation for the work and is relevant to the developments in the following chapters. First, a broad review of the clinical, engineering and commercial literature relating to interface pressure measurement in biomedical applications is provided. This review includes, but is not limited to the three selected target applications: surgical retraction, surgical tourniquets, and mammography. These applications have been selected because they are common procedures and cover a wide spectrum of device and tissue characteristics and properties. Therefore, i f a generic interface pressure transducer could be developed for use in all three different but common applications, it was hypothesized that the same approach could be extended to develop a transducer useful in a broad range biomedical interface pressure applications. The literature for each target application was thus reviewed to define similarities and identify differences that must be addressed when designing a transducer for use in the three applications and in more general biomedical applications. While the measurement of pressure between a device and body tissue is by no means unique to the target applications we have chosen to investigate, only a limited amount of work has been done in the measurement of pressure in these applications. Therefore it is useful to review applications where the measurement of interface pressure has been investigated for a 1 Chapter 2: Background and Review of Previous Research 9 considerable period of time. Many rehabilitation applications involving the prevention of pressure sores, such as the design of wheelchair cushions and bed mattresses, have involved the measurement of interface pressure. Other applications include the measurement of interface pressure on the surface of prosthetic sockets, beneath the foot, and under burn garments. A wide variety of specialized interface pressure measurement techniques have been developed, evaluated and reported in the literature in conjunction with these rehabilitation applications and have been included in the literature review presented here. It was found that the transducers involved in this reported work were not evaluated or calibrated under interface materials of varying compliance. Also, in several cases, curvature of the interface and induced changes to the interface by the transducer adversely affected the transducer output. As the purpose of this research was to develop a generic interface pressure transducer which is both independent of interface material compliance and suitable for use over a wide range of curvatures, none of the reviewed transducer systems were deemed satisfactory. Finally in this chapter, based on the descriptions of the target application areas and the critical review of the interface pressure measurement literature, the design criteria and design specifications for a generic interface pressure have been developed and defined. 2.2 Target applications The target applications were selected because they are common biomedical procedures and encompass a wide spectrum of devices and tissues. Selecting common biomedical Chapter 2: Background and Review of Previous Research 10 procedures is desirable to ensure that the work is significant enough to warrant research. Both interface material compliance and curvature conditions vary widely in surgical retraction, surgical tourniquets, and mammography as described in the following review. 2.2.1 Surgical retractors Almost all surgical procedures require the use of instruments to draw back an impeding soft tissue, organ or bone, to gain access to the surgical site. Even in minimally invasive surgery, parts of the anatomy may need to be drawn out of the way or held in position. The instruments used for this purpose are called retractors and come in many shapes and sizes to accommodate the varying applications. Retractors may be held in place by a variety of methods. A person may hold the retractor, it may be self-retaining, or, it may be held by an automated system [2]. Figure 2.1 shows an example of all three types of retraction approaches. Injuries due to retractors, although not common, have been reported in the literature. Recently a study of back muscle injury in rats due to retractors was conducted by Kawaguchi et. al. [3] who investigated the histology and histochemisty of the retracted muscle over a six-week period. The relationship between retraction pressure, retraction duration and tissue damage was investigated. Although higher pressure and longer times produced more tissue damage, in each case the tissue regenerated itself completely within the six-week period. However, the Figure 2.1 Hand held (top left), self-retaining (top right), and automated retractors (bottom) Chapter 2: Background and Review of Previous Research 12 investigators concluded that muscle retraction injuries are a potential complication of posterior spine surgery which may result in back muscle dysfunction after surgery. In contrast, more serious and potentially irreversible retractor injuries to brain and liver tissue, arteries, and nerves have been reported. Neurologic damage and three deaths were reported by Aserman [4] in 1953 due to high brain retraction pressures which deprived brain tissue of blood and resulted in irreversible neurological damage. Liver damage caused by a retractor during abdominal surgery was reported by Ameli et. al. [5], Injuries due to the compression of the external iliac artery by a retractor which caused ischemia was reported by Lozman et. al. [6]. As well, femoral nerve injury after a gynecological procedure caused by a self-retaining retractor was reported by Schoondorf [7]. These reports demonstrate the potential for serious damage to patients due to surgical retractors. Both the amount of retractor pressure and the duration of the retractor application are important in the incidence of injuries due to retractors. The measurement of retractor pressure has been conducted in neurological procedures and, more recently, in the back muscle study previously mentioned. A wide variety of transducers have been employed in the measurement of the interface pressure developed between the tissue and the retractor blade. Kawaguchi et. al. [3] describe the system used to measure retractor pressure on back muscles as a strain gauge (Kyowa PS-2KA, Tokyo) and a strain meter (Kyowa UCAM-1A, Tokyo). They do not mention calibration techniques or transducer performance characteristics. Chapter 2: Background and Review of Previous Research 13 Several other investigators used strain gauges mounted on various locations of retractors to relate the strain in the retractors to the applied forces in the brain [8-11]. Calibration techniques included hanging known weights from the retractor tip and correlating them to strain gauge output. Although this technique provides quantitative force results, it does not provide localized pressure information since a force applied to an area of the retractor could produce the same output as a lower force applied to a larger area of the retractor. Findlay [2] describes two transducers for measuring the interface pressure in surgical retractor applications. The first system used a saline filled pouch (40 mm x 10 mm x 1 mm) connected to a remote pressure transducer via a saline filled tube. The transducer was calibrated by placing it between a folded blood pressure cuff which was placed between two compression plates. The cuff pressure was monitored and corresponding transducer output measured. The transducer demonstrated some non-linearity and hysteresis characteristics but was subsequently used to measure retractor pressures in total hip replacement surgery. The large surface area of the transducer and the difficulty in obtaining multiple site sensing were stated as major limitations of this device. A second transducer was later developed to overcome these limitations. Findlay's second system incorporated a disposable silicon strain gauge transducer which was embedded in two layers of medical grade silastic sheeting. The completed transducer was 2 mm in thickness and flexible enough to conform to a 40 mm radius of curvature. This transducer was used in 6 abdominal surgery trials to measure retractor pressures. During calibration its input/output characteristics were found to be linear within +/-1 mmHg error when negative Chapter 2: Background and Review of Previous Research 1 4 pressure was applied to the back of the transducer. However input/output characteristics were not noted when tested in the folded cuff configuration which more closely simulates the measurement environment. Strains due to edge effects may have been present under interface conditions adding a component of uncertainty to the output. Calibration of this transducer under interface conditions would be necessary to determine the amount of error due to edge effects. This review indicated that the interface pressure transducers for surgical retractor applications can be classified into two categories: i) strain gauge or ii) pneumatic pouch. Since none of the strain gauge transducers were calibrated under interface conditions, the effect of the interface environment on the transducer outputs was unknown. Although the pneumatic pouch transducer was calibrated under interface conditions, its large surface area was a major limitation. As such, both transducer technologies have limitations which demonstrate the need for a transducer which can be integrated into a surgical retractor to measure the interface pressure between the retractor blade and underlying tissue. Such a pressure transducer could then be used to study the relationship between retractor pressure, application times, and tissue damage. Furthermore, the output of the interface pressure transducer could be used to safely control an automated surgical retractor. Chapter 2: Background and Review of Previous Research 15 2.2.2 Tourniquets Tourniquets are surgical devices that are used during limb surgery to occlude blood flow. The occlusion of blood flow is preferred to limit the amount of blood loss as well as provide a bloodless surgical field. More recently, intravenous regional anesthesia, a technique for local analgesia of the tissues, has required the use of tourniquets to prevent the anesthetic agent from flowing away from the distal surgical site of the limb. Tourniquets are either pneumatic or non-pneumatic devices which occlude blood flow by applying pressure to the circumference of a limb. Figure 2.2 shows an example of a pneumatic tourniquet developed by Western Clinical Engineering Ltd. [12]. Figure 2.2 Tourniquet system Chapter 2: Background and Review of Previous Research 16 < The compression of the underlying tissues and blood vessels results in occlusion of the blood vessels. The amount of pressure required to occlude blood flow is dependent on a variety of factors including cuff design, limb circumference and the mechanical properties of the underlying anatomy. Reports of injury to the tissues underlying a tourniquet due to over-pressurization appear in the literature [13-31], In pneumatic tourniquets, the inflation pressure is monitored in an attempt to control the level of pressure applied by the cuff to the limb. However studies have shown that the pressure distribution on the limb varies with distance from the edges of the cuff. This was shown by McLaren et. al. [32] who conducted experiments to measure the pressure distribution under a standard pneumatic cuff having a width of 85 mm when it was applied to the hind limbs of anesthetized dogs and inflated to a pressure of 200 mmHg. A catheter and fluid pressure transducer were embedded into the tissue at various locations to measure the soft-tissue pressure. The surface pressure followed a bell-shaped distribution. Two important limitations exist with this technique for interface pressure measurements: the technique is invasive; and, the transducer measures hydrostatic fluid pressure. Given that soft tissue behaves as a composite semi-solid material, the relationship between fluid pressure and soft-tissue interface pressure is unclear. In contrast, Breault et. al. [33] used a pneumatic transducer with electrical switches to measure the pressure of tissue underlying a pneumatic cuff in cadavers. Although this technique was used to measure underlying tissue pressures, it could be used for interface pressure Chapter 2: Background and Review of Previous Research 17 measurements as well. However four of the major limitations of this transducer as reported by Breault were: it measured pressures only intermittently; it displayed significant hysteresis; it was unsuited for array measurements; and it was too unreliable for routine clinical use. Non-pneumatic tourniquets are the predecessor of pneumatic tourniquets. With the development of new materials and transducers, it is possible that, in the future, non-pneumatic tourniquets may replace pneumatic tourniquets. Essentially a non-pneumatic tourniquet consists of a bandage which is tightly wrapped around the limb until blood flow is stopped. In contrast to pneumatic tourniquets, non-pneumatic tourniquets do not have any indicators of the pressure applied by the bandage to the limb. An interface pressure transducer could be used to measure the pressure applied by the bandage to the limb. The integration of an interface pressure transducer into a non-pneumatic tourniquet could form the basis for a safe, effective, low-cost tourniquet system. The tourniquet-related literature identifies several techniques for measuring the interface pressure, however, it also demonstrates the limitations of these existing interface pressure transducers. A need exists for an interface pressure transducer which can measure the pressure distribution between a tourniquet cuff and the underlying limb to improve the design, safety and operation of both pneumatic and non-pneumatic tourniquet systems. Chapter 2: Background and Review of Previous Research 18 2.2.3 Mammography Breast cancer is the leading cause of death from cancer for women in British Columbia and Canada [34]. Although there is no known way to prevent breast cancer, early detection of the disease is an important factor in its successful treatment [34]. Mammography is a radiological procedure utilizing x-rays to detect breast cancer. The only other method currently used to detect breast cancer is a clinical breast examination which can be conducted by a physician or by a woman herself. In some cases mammography can detect breast cancer approximately two years before the tumour becomes clinically evident [35]. Although there is much controversy over when mammographic screening should begin and how often screening should occur, mammography is a widespread screening program that had been adopted by many industrialized nations including Canada. In British Columbia, the "Screening Mammography Program" follows the mammography screening guidelines put forth in 1989 by eleven of the world's largest health care and medical research organizations [34]. The guidelines recommend that the screening process begin by age 40 and consist of an annual clinical examination with screening mammography performed at 1 to 2 year intervals. Beginning at age 50, the guidelines further recommend that mammography should be performed on an annual basis. The screening program is fully funded by the B.C. Ministry of Health. The success of detecting breast cancer with mammography depends greatly on the quality of the mammographic x-ray image [36]. With the current technology, compression of the breast is essential to produce the high quality images required by the screening programs. Compression Chapter 2: Background and Review of Previous Research 19 improves the quality of the image by reducing the blurring of tissues within the breast because the tissues are closer to the film. In addition, the radiographic image is sharpened because the breast is effectively immobilized while the effect of scattered radiation which degrades the image quality is reduced. A further benefit of compression is that radiation levels can be reduced due to the reduced thickness of breast tissue [37]. For these reasons, mammography x-ray equipment comes equipped with compression devices. Figure 2.3 shows a typical mammography procedure and an example of the image it produces [38], It is important that there is no slack in the mechanics of the system so that the plates remain parallel, thus leading to a uniform x-ray exposure [39]. Due to variations in both human anatomy (i.e. breast size, shape and composition) and,the mammography x-ray equipment (i.e. the size, shape and tilt of the compression plates), the degree of compression is not the same for each examination. At present, there is no standard regulating this aspect of the mammographic examination procedure. Currently there is no measurement of the pressure directly on the surface of the breast during compression, but rather only a single measurement of the applied force on the compression plates is available on a readout facility. To obtain a more accurate measurement of the interface pressure between the compression plates and the breast tissue, Clark et. al. [1] describe the use of an interface pressure transducer. The system employed consisted of a fluid-filled neonatal cuff connected to a remote Figure 2.3 Mammography procedure and corresponding mammogram Chapter 2: Background and Review of Previous Research 21 pressure transducer via a fluid-filled tube. The authors calibrated the transducer and found the relationship between measured and calculated pressures over the full range of applied forces (0-200 N) to be linear with a correlation coefficient greater than 0.999. Using this transducer, pressure measurements were performed with patients by placing the neonatal cuff between the top compression plate and the patient's breast. The authors state that when the breast was compressed, the cuff remained flat and parallel to the plate. The applied pressure of the compression plates was set to 120 mmHg and engaged. A pressure measurement was then taken from the cuff when the compression plates were applying the set pressure. A total of twelve patients were included in the study yielding 19 readings. The interface pressure measurements ranged from 45.6 to 230.4 mmHg for an applied compression plate pressure of 120 mmHg for different subjects. The results of the measurements clearly show that the applied pressure displayed on the mammography x-ray unit is not well correlated to the interface pressure between the breast and compression plate. The authors considered the reasons for this to be related to both anatomical and mechanical factors. Differences in breast anatomy between patients as well as distortion of the plastic compression plate were given as examples. No mention was made of the possible effects due to the large surface area of the neonatal cuffs, the largest being 126 mm x 54 mm. One would expect a varying pressure distribution between the breast and compression plate, particularly as the plate nears the chest wall. Each patient would likely experience a different pressure distribution on their breast depending on the anatomical factors described. However, the Chapter 2: Background and Review of Previous Research 22 cuff would average these pressure distributions and experience a single uniform pressure for each patient. This averaging process introduces a large degree of error in the measurements. Although this study attempted to gain a more accurate reading than that provided by the mammography x-ray unit, further work is needed to develop an interface pressure transducer array which can measure pressure at more discrete points and thus provide a better representation of the pressure distribution over the surface of the breast. Such a pressure transducer could potentially be used to safely control an automated mammography machine. 2.3 Review of the literature of other biomedical interface pressure measurement applications Several biomedical applications require interface pressure measurements to improve the design, performance, and safety of the devices involved in the applications. It is useful to review the literature of biomedical interface pressure measurement in these applications to assess existing transducers and technologies for use in the target applications previously identified. The following is a review of mattress and cushion applications, prosthetic sockets, foot pressure, and pressure burn garments. Chapter 2: Background and Review of Previous Research 23 2.3.1 Cushion and mattress applications In rehabilitation applications where a person is confined to a wheelchair or bed, pressure relief characteristics of the cushion or mattress is of significant importance. Evaluation of the ability of the cushion or mattress to reduce the magnitude of the contact pressure to delay, if not prevent, the onset of bed sores, is commonly reported both in the scientific and product literature. A wide variety of interface pressure measurement systems were used within this single application area [40-50]. In all of these studies the main objective was to determine the pressure reduction capabilities of specially designed mattresses or cushions. Krouskop et. al. [40] discussed the confusion that exists about the meaning and use of interface pressure measurements reported both in the scientific and product literature. The authors caution that interface pressure data collected by different investigators using different instrumentation are not comparable, stating that factors including size, shape and position of the pressure sensor, as well as instrumentation, effect the value of the pressure measurement. Although no mention was specifically made regarding the effect of interface materials of varying compliance on transducer outputs, this may have also adversely affected transducer outputs. The following examples describe some of the systems used by these investigators and demonstrate their capabilities and limitations. Clark et. al. [42] used a liquid filled 20 mm diameter sensor connected to a pressure transducer (SCP Monitor, Medimatch Ltd.) to measure the contact pressure between the sacrum of groups of young and elderly subjects lying upon mattresses specifically designed to prevent Chapter 2: Background and Review of Previous Research 24 pressure sores. The method of calibration was to place the sensor between a subject's forearm and a sphygmomanometer cuff. The reported accuracy from this method of calibration was +1-3 mmHg below 20 mmHg, +/- 5 mmHg below 80 mmHg and +/- .7 mmHg below 100 mmHg. The authors stated that the sensor was then taped directly over the sacrum during the investigation. However, the difference between calibration conditions and experimental conditions (i.e. the change in interface material compliance) was not noted by the investigators yet it may have contributed to significant inaccuracies in the measurements. In two similar studies Hover et. al. [43] and Petrie et. al. [44] used pneumatically activated switches to measure the interface pressures produced between the mattress and tissue at four body locations: the scapular area; sacral area; trochanter; and heel. Neither study outlines the method of calibration used nor the transducers output characteristics, nor do either group of investigators discuss the potential for inaccurate results due to the differences in the compliance or curvature of the anatomy and mattress within each application. At best, the results show relative levels of pressure indicating high pressure points. As indicated by these examples, there is currently no standardized method of measuring the interface pressure even within one application area. To be able to compare interface pressure measurements, a need exists for an interface pressure transducer which can measure interface pressure independently of the interface material compliance or curvature. Chapter 2: Background and Review of Previous Research 25 2.3.2 Prosthetic sockets Another application where interface pressure measurements have been attempted is in prosthetic devices. The pressure between the socket of a prosthetic device and the limb affects stability, skin condition and patient comfort. Several investigators have developed interface pressure transducers to study prosthetic socket/limb pressure relationships. The adverse effect of varying interface material compliance and curvature on the transducers were noted, although solutions to these difficulties were limited. The following examples describe these interface pressure transducers and demonstrate their capabilities and limitations. In 1969, Appoldt et. al. [51] studied the relationship between socket pressure and pressure transducer protrusion. In this study the transducer (N.Y.U. Pressure Transducer) employed a piston impinging against a stiff spring, sensing pressure as the spring deflected. The deflection was converted into an electrical signal through a network of strain gauges. The transducer was mounted at numerous locations either flush with the interior socket wall or protruding into the socket by 1/16th of an inch. The prosthesis was then donned by subjects with an above knee amputation and pressure measurements were taken during walking strides. Perhaps not surprisingly, a comparison of the measurements from the flush and protruding transducers indicated that in over 42% of the measurements, the protruding transducer produced higher values. In these cases, the ratio of protruding to flush values ranged from 1.5 to 4.0. The investigators found that a high protrusion/flush pressure ratio corresponded Chapter 2: Background and Review of Previous Research 26 to locations of bone or muscle proximity. This experimental finding seemed reasonable when the causes were examined. The investigators discussed a model of the stump as a very compliant material which could be represented as a membrane enclosing a volume of gas. In this case the stump placed within a socket and given a weight-bearing load would be able to distribute the load equally and the flush and protruding transducers would indicate the same pressures. However as the model increased in stiffness, the investigators demonstrated that the stump lost its ability to conform to the socket and the load was re-distributed to. contact points (i.e. to the protruding transducer). In this case the protruding transducer would indicate a higher pressure than the flush transducer. An actual stump resembles a combination of these models. Areas of soft tissue resemble the compliant model while areas of bony protrusions resemble the stiff model. Therefore, this study demonstrates the importance of considering the anatomical characteristics of the measurement environment when choosing the type and location of the transducer. Rae et. al. [52] conducted a study of the basic mechanisms affecting the fit of lower-extremity prostheses with the use of interface pressure transducers. A survey of commercially available interface pressure transducers was conducted which identified two types: 1) a strain-gauge diaphragm operated transducer manufactured by Scientific Advances, Inc. and; 2) a transducer which incorporated a Wheatstone bridge formed directly on a silicon diaphragm manufactured by Kulite Semiconductor Products, Inc.. Both transducers were initially developed for measurements of dynamic surface pressure on helicopter and turbine blades. The investigators installed the transducers in a below-knee prosthesis and state that, although the Chapter 2: Background and Review of Previous Research 27 results indicated the transducers responded reasonably well to the environment, a lack of confidence was reported for use of the transducer on or around bony prominences within the stump. This indicates that the transducer was dependent on the compliance of the interface materials. No indication was provided as to whether or not the transducers were flush with the socket wall nor were any quantitative results reported. Furthermore the pressure vessel which was used to calibrate the transducers did not simulate the interface conditions under which the transducers were being used. Ensuring transducers are flush in prosthetic sockets requires that material be removed from the socket wall, possibly compromising the strength of the prosthesis. In 1980, Van Pijkeren et. al. [53] developed and evaluated a novel transducer for use in stump/socket measurements that was thin enough to be'placed between the stump and the socket wall without socket material removal. The transducer was also flexible and could conform to curved surfaces. Although Van Pijkeren et. al. described the method of calibration, they did not report the transducer calibration curve characteristics. No mention was made of hysteresis, drift or repeatability characteristics. The investigators did report that the transducer was sensitive to bending, stating that it produced an increase in pressure in the liquid of the transducer, but they discounted the significance of the output as long as the curvature is not "exceptional". Unfortunately, the investigators did not define how much curvature they considered to be exceptional. Chapter 2: Background and Review of Previous Research 28 These studies demonstrate the lack of a standard method for measuring interface pressures between limbs and prosthetic sockets. These studies also indicate that issues relating to the differing compliance of interface materials, as well as bending of a transducer, may adversely effect transducer output. 2.3.3 Foot pressure The measurement of pressure distribution beneath the foot is of importance in the investigation of the diabetic and rheumatoid foot, as well as for the assessment of outcomes of certain surgical procedures [54,55]. In a review paper, Lord et. al. [56] briefly described the different techniques of foot pressure measurement including: direct printing; visualization and optical pattern processing; and load cells embedded in force plates or shoe insoles. The investigators mentioned the presence of shear forces which could adversely effect the output of transducers intended to measure normal forces only. The authors also stated that quantitative results from different techniques are difficult to compare. It is probable that issues related to changing interface material compliance and shear forces partially contributed to errors in measurement. In a related article, Roy [57] reviewed force, pressure, and motion measurement in the foot and described criteria for pressure measurement systems as well as measurement techniques. In the discussion of pressure measurement inside the shoe, the author described discrete Chapter 2: Background and Review of Previous Research 29 transducers which can either be attached to the sole of the foot or embedded into the insole of the shoe itself. The major disadvantage of transducers that are not embedded into the insole is that they protrude into the foot, potentially changing the pressure distribution. Even embedding the transducer in an insole could significantly alter forces under the foot depending on the compliance of the insole, again indicating a dependence on material compliance. 2.3.4 Pressure garments for treatment of burn injuries Pressure garments are used extensively to reduce scarring following burn injuries. A wide variety of pressures are achieved when different garments are applied to different parts of the body due to the varying local compliance of both the garment and body tissue. To achieve optimal pressure, an accurate measurement system is required. The following examples describe measurement systems in pressure burn garment applications and demonstrate both their capabilities and limitations. In a recent study, Harries et. al. [58] described the use of an Oxford Pressure Monitor to measure the pressure under burn pressure garments. An accuracy of +/- 5 mmHg was indicated by the suppliers but no mention was made regarding calibration of the transducers before testing. The measured pressures ranged from 0 to 37 mmHg, the authors reporting that in some cases they could not obtain a pressure reading when the transducer pad was placed over soft-tissue areas (i.e. high compliance material). The investigators concluded that although the Chapter 2: Background and Review of Previous Research 30 measurement system enabled them to generally compare pressure ranges between garments over various body parts, a more accurate measurement system would be necessary for incorporation into scar management programs. Barbenel et. al. [59] have developed and evaluated a simple interface pressure transducer for use under pressure garments. The transducer consisted of a flat circular cell constructed of 28 mm diameter, 0.25 mm thick PVC sheets connected to an integrated circuit piezo-resistive, pressure-sensitive device via a 3 mm diameter oil-filled tube. Under hydrostatic calibration, the transducer produced a linear relationship between applied pressure and output, which was statistically confirmed. The transducer was also tested under interface material conditions. It was taped to the forearm and tested under a sphygmomanometer cuff. Under these conditions the transducer responded linearly but with more scatter than for the hydrostatic calibration. In this calibration test, an important limitation was that the transducer was only tested to a maximum pressure of 40 mmHg. The authors also point out that vertical displacement of the transducer cell relative to the remote pressure transducer will produce a hydrostatic pressure which will be measured by the transducer. 2.3.5 Summary of review The following conclusions were reached as a results of a criticaf review of the literature describing interface pressure transducers used in the above applications: the transducers were Chapter 2: Background and Review of Previous Research 31 used in a single application; the transducers were not evaluated or calibrated under interface materials of differing compliance; the curvature of some transducers adversely affected their output; and perturbation of the interface by the transducer altered the pressure distribution. As the purpose of this research was to develop a generic interface pressure transducer for use in the target applications, which is both independent of interface material compliance and suitable for use over a wide range of curvatures, none of the reviewed transducer systems were satisfactory. 2.4 Design criteria and specifications for interface pressure transducers This review of the literature, as described above, provided information and insight on an extensive variety of interface transducers for specific applications. From this review, the capabilities and limitations of many interface pressure transducer systems were identified, and optimal design specifications were developed as described below. 2.4.1 Generic criteria for transducers Regardless of the measurement variable, transducers must demonstrate some common characteristics to be useful. The following characteristics have been identified. Chapter 2: Background and Review of Previous Research 32 1) The measurement must be accurate and repeatable. The degree of accuracy and repeatability required will depend on the specific application. Generally the more accurate and repeatable the measurement system must be, the more costly the system. 2) The transducer must be accurate over the entire range of variable measurement. The transducer must be selected or designed such that its range is slightly larger than the measurement range. It may be difficult to anticipate the maximum value of the variable being measured therefore the transducer should also be able to withstand a variable value that is over that which is expected. 3) The transducer output must remain at the measured value unless the variable itself is changing. If the output is significantly affected by environmental effects (such as temperature or moisture) the output may drift. Therefore the transducer must be compensated for environmental effects. 4) If the transducer is sensitive to disturbances in directions other than that which it is measuring, the output may include a significant spurious component which is referred to as cross-sensitivity. Therefore the transducer must exhibit an insignificant amount of cross-sensitivity. 5) There must be a relationship between the transducer output and the variable to be measured. The transducer output may be linearly related or non-linearly related to the measured variable. Chapter 2: Background and Review of Previous Research 3 3 6) The natural frequency response of the transducer should be high enough to ensure an accurate measurement gradient. Again, speed of response is highly dependent on the application. For example, dynamic foot pressure measurements would require a much faster response than seating pressure measurements. 2.4.2 Specific criteria for an interface pressure transducer As well as satisfying the generic criteria for transducers, an interface pressure transducer must also satisfy criteria specific to the environment it will be operating in. The characteristics of an interface pressure transducer depend on the mechanical properties of the measurement environment, such as the stiffness of the device and body at the interface. These properties vary with each application. For example, in mammography, breast tissue is placed in contact with a flat rigid plate. In tourniquet use, the limb, which is a complex composite of skin, fat, muscle, and bone, is compressed under a curved flexible plastic. However, in surgical retraction, internal tissue is placed in contact with a rigid curved material. Clearly, the anatomy for each application varies significantly and therefore must be addressed in any future interface transducer design. It is our intention to develop a single transducer that will operate equally well in all three target applications, (i.e. applications with varying material compliance) while providing any specific packaging or mounting modifications (as necessary) to accommodate each application. Chapter 2: Background and Review of Previous Research 34 Another requirement is that the perturbing effect of the transducer at the interface of the device and tissue must not significantly effect the transducer output. The perturbing effect of a transducer at the interface is dependent upon the area and thickness of the transducer, its capacity to conform to curved surfaces, and the relative stiffness of the device, transducer, and anatomical structures in the interface region. Ferguson-Pell [60] used an analytical approach to estimate the ratio of area to thickness to be not greater than 10:1. However, the analysis used gross simplifications such as approximating tissue as a linear homogeneous material. Therefore, estimating the maximum thickness and diameter of an interface transducer for a particular application using this analysis would require experimental testing to verify performance. Adding curved surfaces complicates the measurement environment even further. In some applications the interface surfaces may be curved in two dimensions (e.g. spherical) down to a diameter of 20 mm. In these cases the transducer must either be flexible enough to conform to the interface profile or small enough to ensure the whole sensing area is in contact with the loading surface. The transducer output must also remain relatively unaffected by a change in curvature. In a study by Guthrie [61], three air inflated transducers were tested in flat and curved configurations. In this case the addition of curvature did result in a change in transducer output. The authors state that it would be fairly straight forward to measure the transducer output versus radius of curvature. However, in many applications the interface curvature is difficult to determine, therefore the transducer output would be difficult to verify. Chapter 2: Background and Review of Previous Research 35 2.4.3 Functional specifications for an optimal biomedical interface pressure transducer To aid in evaluating both existing and developed transducers, a list of optimal design specifications were identified for comparison with evaluated transducer characteristics as follows: • transducer must measure pressure applied by a specified medical device to a body tissue • transducer must measure pressure in a direction normal to the device/tissue interface • the area of sensing element must be no greater than 1 cm in diameter • the transducer must measure pressure in the range of 1 - 500 rnrnHg • must be capable of being configured in arrays or matrices as required for a given application • transducer output must be independent of the compliance characteristics of the interface materials • transducer output must be independent of the curvature of the interface • transducer and transducer package materials must be biocompatible • transducer and transducer package must be resistant to chemicals encountered in target applications • transducer and transducer package materials must be radiotransparent for imaging applications : • measures with maximum error of +/- 2 mmHg • has maximum hysteresis and/or drift between measurements over a 1-hour time period of+/-2 mmHg Chapter 2: Background and Review of Previous Research 36 • provides an adequate number of measurement data rate for the defined applications • fails in a mode that is obvious to the operator and safe for the patient • must provide temperature compensation over the range of temperatures found in the target applications, typically between room temperature (-21 °C) and body temperature (~37°C) • must comply with relevant IEC, ISO and CSA standards • must not provide any potential electrical or thermal hazards ' • must be immune to levels of electromagnetic interference encountered in target application environments • must be able to withstand sterilization by gamma radiation, ethylene oxide or autoclave • transducer design must permit calibration, or calibration-checking, of the transducer by a clinical user in the target application environment • such calibration-checking must be fast, convenient and intuitive (under 5 minutes) • a reusable transducer intended for integration with instrumentation and long term use should have cost of manufacture, including signal pre-processing hardware, of less than $50 each • a disposable or consumable transducer, intended for integration with "patient-applied parts" which are discarded after a single usage (such as some pneumatic cuffs or retractor blades), should have a cost of manufacture, excluding signal processing hardware, of less than $1 Chapter 2: Background and Review of Previous Research 37 2.5 Summary The objectives of this chapter were twofold: to conduct a comprehensive review of the clinical, engineering, and commercial literature to identify technologies used in biomedical interface pressure measurement, and the limitations of available transducers based on those technologies; and to define optimal specifications for a generic biomedical interface pressure transducer for use in the measurement of pressure at the interface between medical devices and living tissues, organs, and limbs of varying compliance. The review of the target applications of surgical retraction, surgical tourniquets, and mammography revealed a need for the measurement of the interface pressure between each device and the corresponding anatomy. In all three applications, attempts have been made to measure this interface pressure with a variety of measurement systems. However, the systems had major limitations and none would be appropriate for all three applications thus demonstrating the need for a multi-application interface pressure transducer. A further review of the available commercial, engineering and clinical literature relating to interface pressure measurement in biomedical applications revealed a large number of interface pressure measurement systems used in wheelchair cushion, bed mattress, prosthetic socket, and pressure burn garment design, and foot pressure measurement. However, one investigator [40] warned against comparing the data obtained with different measurement systems even within a single application because factors such as transducer size and shape effect the value of the pressure measurement. Another investigator [52] cautioned against using specific transducers around bony prominences Chapter 2: Background and Review of Previous Research 38 indicating a sensitivity of the transducers to the compliance and curvature of the interface materials. Sensitivity of specific transducers to curvature and perturbation of the interface by the presence of the transducers were also noted by investigators [51] as factors causing error in measurements. In summary, the literature revealed the need for a multi-application interface pressure transducer and the lack of an existing transducer to fill this need. To aid in the development of a generic interface pressure transducer, this chapter concludes with a list of developed design criteria and optimal design specifications based on the information and insights gained from the literature review. These criteria and specifications will also be used as a guide in the evaluation of existing and developed transducers in later chapters. To further aid in the development of a generic interface pressure transducer, an understanding of the mechanical response of the system, comprised of the transducer and the interface materials, to applied loads is required. The following chapter discusses the development of a preliminary finite element model which approximates the behaviour of the transducer and interface materials under an applied load. 39 3 FINITE ELEMENT MODEL OF MEASUREMENT ENVIRONMENT AND TRANSDUCER 3.1 Chapter overview The mechanical relationship between an interface pressure transducer and an interface environment under an applied pressure is highly dependent on the geometry and mechanical properties of both the transducer and the interface materials. It.is essential to understand the mechanical response of such a system for a given combination, particularly as it relates to the sensing element of the transducer and any changes in the interface pressure distribution. If the sensing element of the transducer is adversely affected by the interface environment (i.e. responding to stresses that are not due to the normal force it is trying to measure) or if the existing pressure distribution is significantly perturbed by the presence of the transducer, then the transducer output will be unrepresentative of the interface pressure which it is trying to measure. As such, it is imperative to gain some understanding of the mechanical response of the system comprised of a transducer and interface materials under an applied pressure. An approximate numerical method was needed to investigate the transducer and interface material system. The finite element method was chosen to model this system due to its wide spread use and its ability to deal with complexities of the system. A brief overview of the finite element method is provided to get a basic understanding of each step involved in the process. Chapter 3: Finite Element Model of Measurement Environment and Transducer 40 Then these steps were followed as pertaining to the transducer and interface materials under a normal applied pressure. Two physical models of a system comprised of a generic transducer and interface materials were defined for investigation and comparison. These models were selected because they are one of the simplest representations of a biomedical device applying pressure to tissue. These models also represent applications with flush and protruding transducers. In one model the transducer was embedded into the device. In the second model the transducer was placed on the bottom surface of the device and protruded into the soft tissue. The physical models were then divided into geometric models of a finite number of elements, with the appropriate boundary conditions. Once the geometric models were developed, approximating functions were assigned to each element. Then a normal pressure was applied to each device and approximate displacement and stress solutions were obtained. The solutions for each model were compared and, based on these results, several conclusions were drawn pertaining to the design and location of an effective interface pressure transducer. As the main objective of this research was to develop a physical prototype of an interface pressure transducer, the finite element models are intended to be preliminary, however they could be further developed to investigate the effect on the system produced by varying material compliance and behaviours (i.e. visco-elastic versus linear-elastic), modifying the interface curvature or modifying other characteristics of the system in future work. Chapter 3: Finite Element Model of Measurement Environment and Transducer 3.2 Brief overview of the finite element method 41 In many engineering applications it is difficult to find a simple analytical solution to problems that involve geometry or some other feature that is irregular. One way of dealing with this difficulty would be to simplify the problem, however in many cases the solution would then be too inaccurate to be useful. The finite element method is a method of obtaining an approximate numerical solution to a problem [62]. With the current availability of computers that can perform large scale numerical processing, the finite element method is capable of providing a more accurate approximate solution than other less complex methods. Although other numerical approximation methods exist, the finite element method was chosen mainly because it can handle complex geometry. ~~ A field variable (e.g. stress) may possess an infinite number of values throughout a continuum (i.e. a body of matter either solid, liquid or gas) because it is a function of each generic point in the body. The finite element method simplifies a problem by breaking a continuum up into elements and expressing the unknown field variable in terms of assumed approximating functions within each element. In this way, the finite element method turns the problem into one with a finite number of unknowns. Nodes or nodal points can lie on the boundary or in the interior of elements. When they are on the element boundaries, adjacent elements are considered to be connected by them. A field variable has a specific value at the nodes. The approximating functions are defined in terms Chapter 3: Finite Element Model of Measurement Environment and Transducer 42 of the values of the field variable at the nodes. The nodal values of the field variable and the approximating function for the element completely define the behavior of the field variable within the element. Also, the approximating functions are usually chosen so that the field variable or its derivative are continuous across adjoining elements. The nodal values of the field variables are the unknowns in the finite element method representation of a problem. Once they are found, the approximating functions define the field variable throughout the group of elements which make up the model. The finite element method involves the following steps: 1. Discretization of the continuum. The continuum under investigation is divided into a finite number of elements. The elements can be chosen to have different shapes (i.e. triangular, rectangular, etc.) to accommodate the geometry of the problem. All the elements together form the finite element mesh, 2. Selection of approximating functions. Each element is isolated and its properties are defined. The approximating function is chosen to represent the field variable within the element. Often polynomials are used in the construction of the approximating function. 3. Assembly of the element equations to obtain the system equations and solution. The individual element equations, and boundary and loading conditions are combined into one matrix. The matrix of simultaneous equations is then solved for the unknowns. The equations Chapter 3: Finite Element Model of Measurement Environment and Transducer 43 are solved such that at a node, where the elements are interconnected, the value of the field variable is the same. 4. Convergence and error estimation. The finite element representation of the problem is refined unto further refinement no longer significantly affects the solution. This ensures that the finite element model solution is close enough to the actual solution to be useful to the user. Three main sources of error exist in the finite element method: geometric approximations, solution approximations, and numerical computations (i.e. truncated decimals). Steps must be taken to assess the magnitude of these errors and, if necessary, to improve the model until the level of error is acceptable for the particular application. 3.3 Development of the transducer and interface material finite element model The finite element method computer program used for the following analysis was ANSYS [63]. See Appendix A for the program code. The objectives of the investigation were as follows: 1) to define two simple physical models that represent biomedical applications involving a transducer, a device and soft tissue; 2) to develop first approximation finite element models of the physical models; 3) to apply a vertical uniform pressure to the devices; Chapter 3: Finite Element Model of Measurement Environment and Transducer 44 4) to solve for, analyze, and compare the displacement and stress result trends of the models under the applied pressure; and 5) to base conclusions about transducer design and location on the results 3.3.1 Model assumptions Two physical models were developed to represent the simplest examples of a biomedical device applying pressure to tissue as a first approximation. They represent applications with flush and protruding transducers and enabled a comparison of these configurations. The first physical model consisted of a 10 mm diameter x 1 mm thick circular transducer embedded into a 35 mm diameter x 1 mm thick circular device and placed on a 35 mm diameter x 16 mm thick soft tissue which was constrained at the bottom in the y direction (see Figure 3.1). . • y • transducer device tissue ^ x Figure 3.1 Cross-section of the physical model of flush transducer, device and tissue Chapter 3: Finite Element Model of Measurement Environment and Transducer 45 The second model consisted of a 10 mm diameter x 1 mm thick circular transducer placed under a 35 mm radius x 1 mm thick circular device surrounded by a 16 mm thick soft tissue which was also constrained at the bottom in the y direction (see Figure 3.2). The transducer dimensions were chosen based on the work done by Ferguson-Pell [60] who recommended that an interface pressure transducer have a diameter to thickness ratio of no greater than 10 to 1. ! 1 ">x Figure 3.2 Cross-section of the physical model of a protruding transducer, device and tissue Since the systems were axisymmetric, the F E M models were reduced to two dimensional models of the cross-section of the systems. The element shapes were a combination of rectangles and triangles, with smaller elements being used to represent the transducer and surrounding materials such that more accuracy could be obtained in this area which is of the greatest interest. Chapter 3: Finite Element Model of Measurement Environment and Transducer 46 The models were simplified further by placing axis of symmetry boundary conditions at the centres of the systems and modeling only one half of them since their other halves behave identically. As shown in the physical models, the finite element models are also constrained at their bases in the y-direction. The simplified finite element geometrical representations of the systems are shown in Figures 3.3 and 3.4. The element characteristics were chosen from the group of structural two dimensional solid elements available with the ANSYS F E M program because they most closely represented the simplified physical models. The triangular solid and rectangular structural solid were chosen to represent the system and have linear elastic homogeneous properties. The numerical values of Young's modulus and Poisson's ratio for the three component materials represented are given in Table 3.1. The soft tissue modulus was based on values given in the literature [64-68]. The device and transducer were modeled as structural steel. r™™™™~™™™~™"™™"~  Material j Young's Modulus j i 1 i i Poisson's Ratio Device j 150 GPa | j 0.3 | Transducer j 150GPaI i 0.3 | Tissue I 15 kPa | 0.3 Table 3.1 Mechanical properties of the materials Chapter 3: Finite Element Model of Measurement Environment and Transducer O | CO 3 'E 1 « CO Chapter 3: Finite Element Model of Measurement Environment and Transducer Chapter 3: Finite Element Model of Measurement Environment and Transducer 49 As the main objective of this research was to develop and evaluate a physical prototype of an interface pressure transducer, the finite element modeling was intended to be preliminary and developed with the capability to be modified to analyze more complex transducer and interface material models in future work. The simplified model assumptions were as follows in this analysis: • the interface was assumed to be flat • the surface nodes of the device, transducer, and tissue were assumed to be interconnected and unable to slide past one another • the materials were assumed to be linear elastic and homogeneous • a single set of compliance conditions as shown in Table 3.1 were assumed for the materials • the protruding transducer was assumed to be surrounded by tissue (i.e. the gap that would normally be present between the device and tissue caused by the presence of the transducer was eliminated) Although one of the essential design specifications is that the transducer be capable of measuring interface pressure between curved devices and tissues, the issue is minimized in mammography since the interface is not curved. As well, through appropriate packaging and selection of the transducer sensing element size, curvature effects can be minimized in the surgical retraction application. Therefore it was felt that modeling of a curved interface was not essential to the initial development of the interface pressure transducer and was recommended in future finite element model development. Chapter 3: Finite Element Model of Measurement Environment and Transducer 50 A further simplification of the model was that all the materials were interconnected at the nodes on the element boundaries. Although, in reality, thedevice, transducer, and tissue are not one continuous system, the amount of surface friction between these components under an applied pressure could be substantial without any lubricant between them. The F E M model assumes there was no relative movement between the device, transducer or tissue on the element boundaries, and thus provides a worst case scenario for this biomechanical system. The sensitivity of the interface pressure distribution to the perturbing effect of the transducer was a function of both the compliance and behaviour (i.e. linear elastic versus visco-elastic) of the interface materials and the transducer material, and the dimensions of the transducer, device and tissue. Investigating the relationship between these parameters using finite element modeling was beyond the scope of this thesis and has been recommended for future work. Convergence and error estimation were not conducted due to the relative simplicity of the model and the small initial size of the elements chosen (i.e. element sizes ranged 0.5 mm x 0.5 mm to 2.0 mm x 1.0 mm). Finally, eliminating the gap between the device and tissue due to the presence of the protruding transducer by surrounding the transducer with tissue will result in a non-conservative estimate of the stress and displacement solution in this model, however the trends indicated should still be of some value. Chapter 3: Finite Element Model of Measurement Environment and Transducer 51 3.3.2 Model verification The model was verified by: 1) selecting a Young's modulus of 15000 N/m 2 and a Poisson's ratio of 0.3 for all the materials; 2) applying a negative vertical load of 1500 N/m 2 to the device surface; 3) calculating the displacement and stress solution analytically; 4) obtaining the displacement and stress solution through the finite element program; and 5) comparing the analytical and finite element program solutions ^ The complete analysis is found in Appendix A. The analysis demonstrated that the finite element model was functioning correctly, therefore the next steps in the investigation commenced. 3.3.3 Applied loading , finite element solution, and discussion of results The material properties of the respective finite element models were set to those presented in Table 3.1. A uniform vertical pressure of 1500 N/m 2 was applied to each device in the negative y direction (see Figure 3.5). The finite element solutions for both models were then obtained. Chapter 3: Finite Element Model of Measurement Environment and Transducer 52 V * *—i- * * * -0-• 0- 4/ ^  ^ ^ I ^ Figure 3.5 Applied loading 3.3.3.1 Flush transducer model The displacement of the model due to the applied loading met expected results given the model constraints (Figure 3.6). The tissue was generally compressed in the y-direction and expanded at the outer edge in the x-direction, However, since the nodes which are common to the tissue and device or tissue and transducer were interconnected, the tissue was constrained by the x-direction expansion of the device and transducer at this edge. The x-direction stress solution due to the applied loading indicated regions of shear at the interface of the tissue in contact with the device and transducer, and within both the device and transducer (Figure 3.7). The amount of shear stress was a function of the constraint at the this interface. As previously noted, the frictional forces between a device and the tissue it is in contact with in a biomedical application could be quite high, however this finite element model represents the worst-case scenario. Clearly, the presence of shear at this interface should be considered in the development of an interface pressure transducer that is employed to measure normal applied loading only. Depending on the transducer technology used, the potential exists for the transducer to respond to shear stresses thereby not providing an accurate measurement of applied normal loading. Chapter 3: Finite Element Model of Measurement Environment and Transducer 54 Chapter 3: Finite Element Model of Measurement Environment and Transducer 55 The stresses developed in the model in the y-direction indicated that the pressure distribution at the interface of the tissue in contact with the device and transducer was significantly different than the applied pressure loading (Figure 3.8). The constrained tissue placed the device and transducer under a bending moment which has resulted in a tensile stress at this interface, as well as a compressive stress in the top half of the transducer. These developed stresses were partially a function of the tissue constraint at the device and transducer interface and would be reduced if the constraint were relaxed (i.e. if the nodes could slide past one another). These stress solutions illustrated two important points: the interface loading may be significantly different from the applied loading; and a transducer may experience stresses that are not representative of the normal applied loading it is trying to measure. These findings should be considered when developing and evaluating an interface pressure transducer for biomedical applications. 3.3.3.2 Protruding transducer model The displacement of the protruding transducer model due to the applied loading was very similar to that displayed by the flush transducer model except for minor differences in the tissue surrounding the transducer (Figure 3.9). Again, the x-direction stress solution due to applied loading indicated regions of shear stresses at the interface of the tissue in contact with both the Chapter 3: Finite Element Model of Measurement Environment and Transducer 5 b Chapter 3: Finite Element Model of Measurement Environment and Transducer 58 device and the transducer (Figure 3.10). Although the stresses are smaller than those developed in the flush transducer model, methods of reducing shear stresses should still be considered in when using a protruding transducer. The stresses developed in the model in the y-direction indicated that the pressure distribution at the interface of the tissue in contact with the device and transducer was significantly different than the applied pressure loading (Figure 3.11). However, the change was not as great as that produced by the flush transducer model. This finding is somewhat counter-intuitive based on the findings reported in the investigation comparing flush and protruding transducers in prosthetic sockets (Chapter 2) [51]. However, as previously noted, this model simplified the actual measurement interface by eliminating the gap between the device and the tissue due to the presence of a transducer. This simplification resulted in a non-conservative estimate of the stresses developed in the model due to the applied loading. As such, further finite element modeling which can be verified by physical testing is required compare the performance flush and protruding transducers. 3.4 Summary This chapter described the development of two preliminary finite element models of systems comprising of a transducer and interface materials to aid in understanding the mechanical response of these systems to an applied pressure. One model investigated a flush transducer and Chapter 3: Finite Element Model of Measurement Environment and Transducer 5 9 - L -z CTl r-O H OJ ^ n H I I I I I I H (N LO CO IIIII10II u o CO 3 ea 60 g -a o •— P. u — o 60 g — o a. p. U 3 a O — X 8) M .3 <D en u '— 3 SO Chapter 3: Finite Element Model oj Measurement Environment and Transducer Chapter 3: Finite Element Model of Measurement Environment and Transducer 61 the other model a protruding transducer. Several important observations were noted regarding the mechanical response of these systems. First, the transducer experienced shear stresses in both models. Depending on the transducer technology used and the type of sensing element, a shear stress could adversely effect the transducer output. Secondly, the pressure distribution at the interface is significantly different than the applied loading. These trends in the finite element solution need to be considered when developing an interface pressure transducer. In the following chapter, a variety of transducers were obtained or developed for evaluation as interface pressure transducers. The observations made from this preliminary finite element model analysis aided in the understanding of these transducers' responses under interface conditions. 62 4 D E V E L O P M E N T O F A C A L I B R A T I O N S Y S T E M A N D E V A L U A T I O N O F I N T E R F A C E PRESSURE T R A N S D U C E R S 4.1 Chapter overview An objective of this research was the development of a suitable calibration system for the evaluation and comparison of a variety of interface pressure transducers. Ferguson-Pell [60] emphasized the critical importance of using a calibration technique that simulates interface conditions when calibrating an interface pressure transducer. The calibration system developed for this research was based on a design described by Ferguson-Pell and was adapted and improved to allow testing under interface materials of varying compliance. Testing transducers under different interface curvature conditions was not performed at this stage for the reasons stated in Section 3.1, however the calibration system could incorporate curved fixtures to simulate this condition. During construction of this system, an extensive survey of prototype and commercially available transducers was conducted to determine which transducers had potential for use in the target applications. The identified transducers were then obtained or constructed for evaluation in the calibration system. In total, nine transducers were evaluated in the calibration system. These transducers were chosen because they represented a wide variety of transducer technologies and had relatively thin profiles to accommodate the dimensional constraints of an Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 63 interface pressure transducer. These transducers were tested for hysteresis, drift, and sensitivity to interface material compliance and their characteristics were compared with the optimal design specifications (Section 2.4.3). None of the transducers tested were satisfactory for use in all three target applications in their current configuration and packaging. However, based on this evaluation, the transducers were prioritized for potential further development and common performance characteristics were noted which lead to the development of essential criteria for an effective generic interface pressure transducer. 4.2 Calibration system A calibration system which simulates the measurement environment is critical for evaluating transducers for interface pressure measurements. As previously mentioned, the measurement environment consists of a transducer placed between a device and a tissue (Figure 4.1). The properties of both the device and tissue change significantly with application, however it is reasonable to assume that these materials exhibit three dimensional stresses and strains and are not well modeled by a hydrostatic fluid. The stiffness characteristics of interface materials may adversely effect transducer output due to edge effects when a transducer is placed between these materials. Therefore the use of a pressure chamber (which provides an interface comprising of a device and fluid) to calibrate and evaluate transducers for use in an interface comprising a device and tissue is not recommended. Chapter 4: Development of a Calibration System and Evaluation ofInterface Pressure Transducers 64 Device Applied Pressure Figure 4.1 Schematic of the interface measurement environment Several methods have been used to simulate an interface environment including compression plates with known applied weights, compression plates applied to a folded over blood pressure cuff, and a pressure chamber applied to the top layer of two interface materials. The most versatile and consistent of these three methods is the pressure chamber applied to the top layer of two interface materials. A specific design of a pressure chamber applied to the top layer of two interface materials calibration system was reported by Ferguson-Pell [60]. The system was described as being a clear plastic pressure chamber, a rubber membrane, a sealing ring, and a base plate which were aligned and bolted together (Figure 4.2). A slot in the sealing ring was designed to allow the pressure transducer to be placed between the rubber membrane and the base plate. The pressure chamber was inflated to a known pressure thus loading the pressure transducer in a direction Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 65 perpendicular to the interface surface. A sphygmomanometer was used for inflation and measurement of this applied pressure. A pressure range of 0-300 mmHg was recommended. Sphygmomanometer Figure 4.2 Calibration system designed by Sachs The calibration system designed and developed for use in this research project was based on the system designed by Sachs with modifications incorporated to increase the maximum allowable pressure and automate the air source and pressure sensing mechanism ( Figure 4.3). All components, except for the membrane material and o-ring, were made of aluminum. An o-ring was incorporated into the design to increase the strength of the seal between the pressure chamber and the membrane material to increase the allowable pressure in the chamber. An ATS 1500 (Zimmer, USA) tourniquet system was used as the air source and pressure sensing mechanism. This system allowed presetting to a desired pressure and pressure incrementing in 1 Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 66 mmHg intervals. The accuracy of this system was specified to be +/- 2 mmHg by the manufacturers. The calibration system has an area approximately 16 cm in diameter and can accommodate a circular piece of material up to approximately 10 mm in thickness. An essential capability of the developed calibration system is that it allows the testing of different materials to determine what effect, if any, they may have on the transducer output. Figure 4.3 Components of the developed calibration system 4.3 Evaluation of existing interface transducers One of the objectives of this research was to select appropriate transducers and evaluate their capabilities in the calibration system with a variety of interface materials of varying compliance. The purpose of this section is two-fold: Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 67 i) to determine the input/output characteristics of a variety of transducers placed under a variety of interface materials and; ii) to compare the output of transducers and prioritize them according to their ability to provide interface pressure measurements under interface materials of varying compliance and their potential for further development. The following sections describe the experimental design and the transducers tested, followed by a discussion of the results. 4.3.1 Experimental design The transducers were tested under several conditions to quantify the measured relationship between applied pressure and transducer output, hysteresis and drift characteristics, and the effects of interface materials having different compliances. The basic experimental design was to test each transducer between the aluminum base plate and a latex membrane (approximately 1 mm thick) and was termed as test condition 1 (Figure 4.4). Under these conditions, the applied pressure was incrementally increased from 0 to 450 mmHg and held for 15 minutes to assess drift characteristics. The pressure was then incrementally decreased back down to 0 mmHg. At each incremental pressure level, the transducer output was recorded. Chapter 4: Development of a Calibration System and Evaluation ofInterface Pressure Transducers 68 Depending on their performance under test condition 1, several transducers were also tested for their performance characteristics between varying interface materials. Various transducer orientations and material lay-ups were used in this testing (Figure 4.4). The interface materials consisted of urethane gels of varying compliance. 4.3.2 Transducers The transducers considered in this work were selected from sources including the commercial, engineering, and clinical literature and cover a wide range of transducer technologies. The following section describes the operation and design of each of the transducers tested and evaluated. 4.3.2.1 Transducers based on force sensitive resistors The Uniforce^^ transducer (Force Imaging Technologies, Chicago, IL, USA) is constructed having two layers of conductive material with two layers of pressure sensitive material in between. The standard transducer is approximately 0.075 mm thick and the sensing element is 6.35 mm in diameter (Figure 4.5) however a variety of configurations and arrays are available. As pressure is applied to the sensing area, the two layers of pressure sensitive material Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 69 Legend Latex - L Urethane - U Transducer - T Sensing element up - Su Base plate - B Sensing element down - Sd ^ . r T, Su test condition 1 U _ , T, Su £= mmmmmmm test condition 2 L T, S d ^ \ B. test condition 3 test condition 4 Figure 4.4 Transducer orientations and interface material lay-ups used in the calibration system Chapter 4: Development of a Calibration System and Evaluation ofInterface Pressure Transducers 70 compress together and cause a decrease in resistance which is related to the pressure on the sensing element. The transducer is available with the Uniforce Experimenter's Ki t which includes the hardware and software necessary to operate the transducer from a computer. The cost of a single transducer was approximately $15 and the U n i f o r c e ^ ^ Experimenter's K i t was around $600 at the time of purchase. For a complete description of the transducer construction and characteristics, refer to the Uniforce Technical Note #101 (2/94) [69]. A second force sensitive resistive transducer is the F S R (Interlink, Santa Barbara, C A , U S A ) which is constructed from a layer of conductive polymer, an air gap open to atmosphere, and another layer of conductive polymer with a pattern of interdigitated fingers (Figure 4.6) [70]. The transducer sensing element is 20 mm x 20 mm and the transducer package is less than 1 mm thick. They are also available in a variety of other sizes and configurations. As pressure is applied to the sensing area, the two conductive layers are compressed together and the resistance between the two sets of fingers decreases. The cost for a stock transducer is approximately $1. Dimensions in inches Figure 4.5 Uniforce transducer schematic T M Chapter 4: Development of a Calibration System and Evaluation ofInterface Pressure Transducers 71 Conductive polymer Mylar backing Output Figure 4.5 F S R / ^ transducer schematic - unfolded view 4.3.2.2 Capacitive transducer A capacitive type transducer prototype (Nam Tai Electronics, Burnaby, B.C.) is constructed from two parallel 7 mm x 7 mm metal plates spaced approximately 1 mm apart. Accompanying micro-circuitry is necessary to detect changes in capacitance with changes in plate distance. The change in capacitance is determined by the change in the frequency response. The transducer requires a 3 volt power supply and a frequency counter to operate. As pressure is applied, the distance between the plates reduces and the frequency decreases. The sensing element and micro-circuitry is embedded into a flexible silicone tapered shim to reduce the edge effects. The overall shim is approximately 50 mm in diameter x 5 mm thick at the centre, tapering down to 1 mm at the edges. Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 72 4.3.2.3 Intracranial pressure transducer with strain gauged diaphragm The Mikro-tip pressure transducer (Millar Instruments, Houston Texas, USA) sensing element dimensions are 1 mm x 2 mm and overall dimensions are 2 mm x 4 mm with a 2 mm diameter cable connecting the transducer to the instrumentation. The strain gauges are in a balanced Wheatstone bridge configuration and as the diaphragm deflects, the resistance changes, which unbalances the bridge, causing an output voltage to be produced. A 5 volt power supply was used for excitation and a voltmeter was used to measure the bridge voltage output. 4.3.2.4 Transducers having constant pneumatic flow A prototype transducer was constructed based on the design of an intracranial pressure transducer reported in the product literature [71]. The transducer consists of a circular base with a plenum and central ring. An intake port is drilled into the plenum and an exit port is drilled into the central ring of the transducer. Hoses are connected to the intake and exit ports. A diaphragm is bonded to the chamber such that it is flush with the central ring (Figure 4.7). The prototype transducer sensing element is 14 mm in diameter. The overall transducer package is 21 mm in diameter x 6 mm thick. The base is constructed from aluminum and the diaphragm is made of mylar. The diaphragm is bonded to the chamber with a methyl-2-cyanoacrylate adhesive. Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 73 To obtain a pressure reading with this transducer, a known flow of air is delivered to the chamber via the intake hose. It fills the plenum and applies pressure to the diaphragm, allowing the air to escape through the exit port in the centre of the transducer. As pressure is applied to the diaphragm, the amount of resistance required for the air to exit at constant flow increases. This resistance is overcome by an increase in the intake air pressure which can be measured and related to the applied diaphragm pressure. Air to Monitoring Unit Diaphragm Plenum Exhaust Tube Air enters plenum through the inlet tube (not shown) and leaves through the exhaust tube Figure 4.7 Constant pneumatic flow transducer schematic A Steritek® intracranial pressure transducer (Steritek Devices, Moonachie, NJ, USA) operates on the same principle as the prototype previously described. However its sensing element is only 7 mm in diameter while the overall package is 10 mm in diameter x 2 mm thick (Figure 4.8) [72]. A flow meter is required to ensure a constant flow rate and a manometer is required to measure the intake air pressure. I Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 74 Figure 4.8 Steritek® transducer schematic 4.3.2.5 Transducers employing fluid-filled pads with remote fluid pressure measurement The first transducer considered was an oil filled 30 mm diameter pad connected to a remote fluid pressure transducer via an oil filled tube. The transducer is based on a design reported in the literature for use in interface pressure measurements [59]. A similar transducer was constructed based on another design reported in the literature and a description of the construction technique is given in reference [53] (Figure 4.9). The sensing element is 6 mm in diameter. Oil is used to fill the sensing element pouch, tubing and remote fluid pressure transducer. The basis of operation of both these transducers is that as pressure is applied to the sensing pad, the pressure is transmitted via the oil filled tube to the remote fluid pressure transducer and a measure of the applied pressure is obtained. Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 75 !Isensor disk; I \oil filled tube pressure transducer externally attached to socket wall. signal conditioning socket wall Figure 4.9 Oil-filled transducer system schematic The last transducer considered of this type was a prototype (Western Clinical Engineering Ltd, Vancouver, BC, Canada) pneumatic rectangular bag with a linear array of internal electric switches approximately 25 mm apart (Figure 4.10). In this design, the applied pressure is equivalent to the internal inflation pressure required to open the internal electric switches. Electrical Connector ' Figure 4.10 Pneumatic electric switch transducer schematic Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 76 4.3.3 Test procedure, results and discussion Each of the identified transducers were tested in the calibration system under test condition 1 (Section 4.3.1). Based on the results of the initial testing, several transducers were tested further under test conditions 2 through 4 (Section 4.3.1). The transducer evaluation results and a discussion of the effectiveness of each transducer as an interface pressure transducer are described next. 4.3.3.1 Transducers based on force sensitive resistors The measured relationship between applied pressure and voltage output under test condition 1 for the Uniforce™. transducer indicated both excessive drift and undesirable hysteresis characteristics (Figure 4.11). A static pressure of 450 mmHg applied for 15 minutes caused the output to increase by 26.9%. This level of drift is unacceptable for the target applications which require an accurate measurement of static pressure over the length of a medical procedure. The measured relationship between the applied pressure and the FSR™ transducer output under test condition 1 also demonstrated drift and hysteresis levels which are unacceptable for the target applications. Furthermore, these transducers were found to be sensitive to curvature (i.e. producing output when curved) even with no applied pressure. These transducers were not tested further due to these inherent limitations. Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 11 Applied Pressure (mmHg) Figure 4.11 Input/output characteristics of the Uniforce^^ transducer for test condition 1 4.3.3.2 Capacitive transducer The measured relationship between applied pressure and output frequency under test condition 1 for the Nam Tai capacitive transducer was found to be nearly linear with no significant hysteresis or drift (Figure 4.12). Based on these results, the transducer was tested further in a number of orientations with various interface materials. In one case under test condition 2, the output drifted during the 15 minute static pressure application (Figure 4.13). The measured relationship between the applied pressure and the transducer output frequency under Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 78 test conditions 3 and 4 demonstrated saturation at approximately 250 mmHg which is indicated by the output frequency dropping to 0 Hz (Figure 4.14). g 300 --CT1 250 --£ 200 --I 1 5 0 -8 1 0 0 " 50 --0 50 100 150 200 250 300 350 400 450 Applied Pressure (mmHg) Figure 4.12 Input/output characteristics of the Nam Tai transducer for test condition 1 —x—'Test condition 2 \ -n- test condition 2 - repeat £ 300 --i H 200 -| 100 -0 -I 1 1 1 1 1 1 1 1 1 0 50 100 150 200 250 300 350 400 450 Applied Pressure (mmHg) Figure 4.13 Input/output characteristics of the Nam Tai transducer for test condition 2 Chapter 4: Development of a Calibration System and Evaluation ofInterface Pressure Transducers 7 9 The results of these tests demonstrate the following characteristics of the Nam Tai transducer: a) it is sensitive to the compliance of the interface materials; b) shear forces on the surface of the sensing element may be affecting the transducer output; and c) it is susceptible to drifting during periods of static pressure. In its current configuration, this transducer would not be suitable for applications involving two compliant interface materials such as tourniquets. Also, the non-radiotransparent characteristic of the capacitive plates is unsuitable for mammography applications due to the artifact that would be apparent on the x-ray from the transducer. However the transducer could possibly be developed further for use in a surgical retraction application. It is no greater than 10 mm in diameter, is capable of measuring pressure in the range of 0 - 500 mmHg and could be configured into a linear or rectangular array. As well, it is feasible that it could meet most of the specifications for physical characteristics, Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 80 accuracy, reliability, reproducibility, calibration, cost, electrical characteristics, and sterilization with further development. 4.3.3.3 Intracranial pressure transducer with strain gauged diaphragm The measured relationship between applied pressure and output voltage under test condition 1 for the Mikro-tip pressure transducer indicated linear input/output characteristics plus minimal hysteresis or drift (Figure 4.15). Based on these results, the transducer was tested further in a number of orientations with various interface materials. 9 T 8 + 9 7 ^ itage (m itage (m 6 -itage (m 5 -o > 4 --3 -5—H + - » .—» O 2 — 1 -0 -0 50 100 150 200 250 300 350 400 450 Applied Pressure (mmHg) Figure 4.15 Input/output characteristics of the Mikro-tip transducer for test condition 1 Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 81 The measured relationships between the applied pressure and the transducer output voltage for test conditions 2, 3, and 4 demonstrate that the transducer output was affected considerably by the compliance of the material placed against the sensing element (Figure 4.16). A much higher output/input ratio was obtained when the sensing element acted against the stiffer urethane gel. This can be explained by the fact that the latex membrane can conform around the small transducer radius whereas the more rigid urethane cannot conform to the same degree. Therefore, more of the load is shed to the transducer on the side touching the urethane producing a higher pressure reading. Furthermore, although the transducer is small, its cylindrical shape causes some degree of perturbation of the interface materials which is dependent on their compliance characteristics. Although this transducer does meet our optimal specifications for bio-compatibility and sterilization, in its current configuration and packaging, it is not suitable for use as a generic interface pressure transducer due to its dependence on interface material compliance. However, re-designing the packaging to minimize the perturbation of the interface materials could potentially alleviate some of the transducer's limitations. 4.3.3.4 Transducers with constant pneumatic flow The measured relationship between the applied pressure and the prototype constant pneumatic flow transducer (Western Clinical Engineering Ltd, Vancouver, BC, Canada) output for test condition 1 demonstrated that the input/output characteristics were approximately linear Chapter 4: Development of a Calibration System and Evaluation ofInterface Pressure Transducers 82 test condition 2 ! —x— test condition 3 —o— test condition 4 70 T > S > o 0 0 100 200 300 400 500 Applied Pressure (mmHg) Figure 4.16 Input/output characteristics of the Mikro-tip transducer for test conditions 2,3 & 4 with minimal hysteresis or drift (Figure 4.17). Based on these results, the transducer was tested further using test conditions 3 and 4 (Figure 4.17). These curves show that the input/output ratios demonstrated a sensitivity to the compliance of the interface materials. This sensitivity can be explained by examining the action of the diaphragm of the transducer. For this transducer to operate, its diaphragm deflects outwards, deforming the material it is acting against so that the air can escape through the exit hole at the centre of the transducer. As the material against the diaphragm becomes stiffer, its capacity to deform decreases, and the diaphragm must exert more force to deflect. The interface material may only be in contact with the diaphragm in the centre thereby concentrating the load and causing a lower input/output ratio. In its current configuration, this transducer is not suitable for use as a generic interface pressure transducer. Due to the nature of this transducer's limitations, no further development is recommended. Chapter 4: Development of a Calibration System and Evaluation ofInterface Pressure Transducers 83 Applied Pressure (mmHg) Figure 4.17 Input/output characteristics for the constant pneumatic flow transducer prototype for test conditions 1,3, and 4 The Steritek® transducer also demonstrated compliance-dependent behaviour for test conditions 1 and 2 (Figure 4.18). This transducer also demonstrated no change in output until more than 50 mmHg of external pressure was applied. Therefore, in its current configuration, this transducer is not suitable for use as a generic interface pressure transducer and no further development is recommended. 4.3.3.5 Transducers employing fluid-filled pads with remote fluid pressure measurement The measured relationship between the applied pressure and the output pressure for the oil-filled pad for test condition 1 demonstrated approximately linear characteristics with minimal hysteresis and drift (Figure 4.19). Based on these results, the transducer was tested further using Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 84 — x _ test condition 1 —o— t^est condition 2 \—n— test condition 2 (stiffer gel) Applied Pressure (mmHg) Figure 4.18 Input/output characteristics of the Steritek® transducer test conditions 2 and 4 (Figure 4.19). Unfortunately this transducer also demonstrated a sensitivity to the compliance of the interface materials. This sensitivity may be caused by the inability of the stiffer interface materials to completely conform around the transducer. In these cases, the interface materials are in contact with a smaller surface area of the transducer, shedding more load onto a smaller surface area. The pressure inside the transducer pad increases accordingly and a higher output/input ratio results for the equivalent applied pressure. The smaller oil filled transducer demonstrated non-repeatable, non-linear input/output characteristics when tested using test condition 1 (Figure 4.20). No further testing was conducted on this transducer due to its limitations. Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers test condition 1 V-x—/test condition 2 ~ D — test condition 3 85 20 40 60 80 . 100 Applied Pressure (mmHg) 120 140 160 Figure 4.19 Input/output characteristics of the oil-filled pad for test conditions 1, 2, & 3 120 j 100 --B, 80 --isure 60 -cu >-40 -3 OH 20 -"3 o 0 P -20 300 Applied Pressure (mmHg) Figure 4.20 Input/output characteristics of the smaller oil-filled pad for test condition 1 Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 86 The input/output characteristics of the pneumatic, electric-switch transducer for test conditions 1 and 2 demonstrate a sensitivity to the compliance of interface materials, possibly due to the inability of these materials to completely conform to the shape of the inflated transducer (Figure 4.21). :/ test condition 1 • test condition 2 —x— test condition 2 (stiffer gel) 50 100 150 200 250 300 Applied Pressure (mmHg) 350 400 450 Figure 4.21 Input/output characteristics of the pneumatic electric switch transducer for test conditions 1 and 2 In their current configuration, these transducers are not suitable for use as a generic interface pressure transducer. Although these transducers meet many of the optimal interface pressure transducer specifications, no further development of this transducer type is recommended due to the nature of its limitations. Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 87 4.4 Conclusions A comparison of the nine transducers tested indicate that none are suitable for use in all three target applications in their current configuration and packaging (Table 4.1). Through the evaluation of these transducers, an essential characteristic for a compliance-independent interface pressure transducer has become apparent. Transducers which operate by relating applied pressure to the deflection of their sensing element are highly dependent on interface material compliance. As such, in theory, if the deflection of the sensing element could be minimized, the compliance-dependence would also be minimized. Subsequent transducer development utilized this idea in the development of a new prototype device as will be discussed in Chapter 5. Transducers Drift Hysteresis Compliance-dependent Curvature-dependent Uniforce™ yes yes n/a yes FSR™ yes yes n/a yes Nam Tai intermittent yes yes n/a Mikro-tip no no yes n/a constant pneumatic flow no no yes n/a Steritek® no no yes n/a oil-filled no no yes n/a smaller oil-filled yes yes n/a n/a pneumatic, electrical switch no no yes n/a Table 4.1 Comparison of transducer tested Chapter 4: Development of a Calibration System and Evaluation of Interface Pressure Transducers 88 4.5 Summary Nine transducers representing five transducer technologies were identified, and obtained or constructed for evaluation as interface pressure transducers. The transducers were tested in the developed calibration system under several interface materials of differing material compliance. The results of the evaluation showed that none of the transducers tested were satisfactory as interface pressure transducers in all three target applications in their existing configuration and packaging. The capacitive transducer and strain-gauged diaphragm transducer were the only transducers which were identified as being, suitable for further development in an attempt to overcome their current limitations for interface pressure measurements. The transducer evaluations led to further development of optimal specifications for a generic interface pressure transducer for biomedical usage. As stated in the design specifications (Section 2.4.3), an interface pressure transducer must be independent of the compliance of the interface materials. During transducer evaluations it became apparent that transducers which estimated applied pressure based on the deflection of their sensing elements were highly dependent on the compliance of the interface materials. In theory, if a method of measuring the applied pressure could be achieved without deflection of the sensing element, dependence on the interface material compliance could be minimized or eliminated. As such, further development of a generic interface pressure transducer involved the development of this essential characteristic as described in Chapter 5. 89 5 D E V E L O P M E N T O F A N O V E L I N T E R F A C E PRESSURE T R A N S D U C E R 5.1 Chapter overview As stated in the thesis objectives, one of the main goals of this research was to develop and evaluate a transducer that can accurately measure the pressure at the interface of a biomedical device and tissue. Based oh the conclusions drawn from both the finite element modeling (Chapter 3) and the evaluation of selected transducers (Chapter 4), a transducer technology was selected for development. During transducer evaluations it was found that transducers which related applied pressure to sensing element deflection were highly dependent on the compliance of the interface materials. Since compliance-independence is an essential characteristics of a generic biomedical interface pressure transducer, it was decided to pursue a transducer technology which minimized sensing element deflection. The developed transducer was based on the concept that, by balancing interface pressure on a diaphragm with a fluid pressure applied to the bottom of the diaphragm, a measure of the average applied pressure could be obtained by measuring the applied fluid pressure. An initial transducer was constructed to test the concept. Improvements were incorporated into a second generation prototype to improve the sensitivity of the transducer and reduce the radio-opacity of the initial design. A third prototype was re-packaged and evaluated for use under tourniquet cuffs. The transducer was then compared to the optimal design specifications and found to be satisfactory for incorporation into target clinical applications for testing. s Chapter 5: Development of a Novel Interface Pressure Transducer 90 5.2 Underlying principle of transducer design A transducer which senses pressure (i.e. the normal component of force) by the amount of deflection of its diaphragm will be highly dependent on the stress-strain or compliance characteristics of the interface material. If the interface material is very compliant, it will conform to the diaphragm much like a fluid, and the contact interface pressure will more closely represent applied pressure (Figure 5.1). However, if the interface material is rigid, it will be unable to conform to the diaphragm. In this case, the pressure will be shed from the diaphragm to the solid material around the diaphragm which will develop stresses until it has reached a state of equilibrium. If this occurs, the diaphragm deflection will not correlate well to the applied pressure (Figure 5.1). Applied Pressure HCM MM 111111 L C M t t m m w 4 / / / / / Load Transfer Through Material to Transducer Figure 5.1 Load transfer through material to-transducer Chapter 5: Development of a Novel Interface Pressure Transducer 91 A method of minimizing the deflection of the sensing element of a transducer, while maintaining a method of measuring the interface pressure, is to apply fluid pressure to the bottom of the diaphragm. Theoretically, without the presence of any shear, bending or torsional forces (i.e. pressure is distributed uniformly), the interface pressure applied to the diaphragm would be indicated by the amount of fluid pressure required to maintain the diaphragm in its original position. Although in practice the applied interface pressure may not be uniformly distributed, the resultant effect could be minimized by employing a small diameter diaphragm (i.e. recall original specifications state that the transducer sensing area must be less than or equal to 10 mm in diameter). 5.3 Transducer operating principles, components and instrumentation The operating principles, components and instrumentation described in the following sections apply to all the developed prototypes. The method of detecting the position of the diaphragm is also described. 5.3.1 Transducer operation The transducer operates by balancing interface pressure on a diaphragm with an equal and opposite fluid pressure. The fluid pressure can be measured using any number of effective, inexpensive fluid pressure transducers. The fluid pressure is theoretically equivalent to the Chapter 5: Development of a Novel Interface Pressure Transducer 92 applied interface pressure if the diaphragm is returned to its original, unloaded position. The applied interface pressure will be averaged by the opposing fluid pressure over the area of the diaphragm. 5.3.2 Transducer components Many methods could be employed to measure the position of the diaphragm to determine when it has returned to its original position. Examples include measuring the capacitance between the diaphragm and a base plate, the current through an electrical contact, the intensity of light bouncing off the diaphragm using fibre-optics, or the amount of strain in a strain gauge mounted to the diaphragm. Metallic plates used in some capacitive type transducers would leave an unacceptable artifact on a mammographic x-ray image. Electrical switches would only provide an electric signal intermittently which could potentially complicate the control of a transducer. Fibre-optic technology, although appealing due to the lack of electricity and radio-opaque materials in the transducer, is much more complicated and costly than simple strain gauge technology. Therefore, to test the concept, a simple strain gauged diaphragm is employed here to measure the position of the diaphragm. To measure the point of zero deflection, a single strain gauge is mounted to the diaphragm and utilized as a single arm in a balanced Wheatstone bridge circuit (Figure 5.2). When the diaphragm deflects, the resistance in the strain gauge changes which unbalances the Chapter 5: Development of a Novel Interface Pressure Transducer 93. bridge and causes a voltage output. This principle can be utilized to determine when the diaphragm is returned to its original position by monitoring when the voltage output from the bridge returns to zero. Figure 5.2 Wheatstone bridge circuit 5.3.3 Instrumentation The instrumentation required to operate the transducer includes an air source and pressure sensing device to pressurize and measure the chamber pressure, and a voltmeter to measure the Wheatstone bridge voltage output. The ATS 1500 (Zimmer, USA) was used as the air source and pressure sensing device. The Hewlett Packard 783 53B patient monitor was utilized to monitor the Wheatstone bridge voltage output. The patient monitor was used because of its dual analog and digital display and zeroing capabilities. The chamber pressure was Chapter 5: Development of a Novel Interface Pressure Transducer 94 controlled manually, however hardware and software could be developed to auto-regulate this function. 5.4 Transducer prototypes The current physical embodiment of this novel interface pressure transducer was developed through several design and material iterations. The following is a description of the improvements that were made to develop a transducer that met many of the optimal specifications. 5.4.1 Proof-of-concept prototype An initial simple prototype was developed to assess the feasibility of developing the balanced-diaphragm transducer concept further. A 12 mm diameter and 3 mm depth circular chamber was drilled into a rectangular piece of high density polyethylene approximately 50 mm x 50 mm x 6 mm thick. Then a 3 mm diameter air port was drilled through the thickness of the polyethylene into the chamber and tubing was force fit into the port. The strain-gauged diaphragm consisted of a 390 Ohm strain gauge resistor bonded to a piece of stainless steel shim stock. Wire (30 gauge) was then soldered to the strain gauge tabs! The diaphragm was then bonded to the base using film transfer adhesive such that the strain gauge was directly over the Chapter 5: Development of a Novel Interface Pressure Transducer 95 centre of the chamber opening. Next the strain gauge was bonded to the top surface of the. diaphragm and a three-quarter Wheatstone bridge was constructed from 390 Ohm resistors which were connected to the strain gauge to complete the full bridge configuration. A 5 volt DC power supply was used to excite the bridge. The prototype transducer was tested in the calibration system using test condition 1 (Section 4.3.1) in two modes of operation, first without equalizing back pressure (i.e. measuring the change in bridge voltage output versus applied pressure) and subsequently utilizing back pressure (i.e. measuring back pressure required to return the diaphragm to its original position under the applied loading). The measured relationship between the applied loading and the transducer voltage output without back pressure was found to be non-linear (Figure 5.3). In the subsequent testing under conditions 1 and 2 with equalizing back pressure, the transducer voltage output was monitored and when the it returned to zero, the back pressure was recorded and plotted (Figure 5.4). The curve shows the input/output characteristics to be approximately linear with minimal drift and hysteresis and a repeatable offset of approximately 25 mmHg. - A significant characteristic of this transducer's output is that it is not significantly dependent on the compliance of the interface materials tested. Based on the results of the initial development and evaluation of the balanced-diaphragm transducer, a second generation prototype was developed to address some of the optimal specifications unmet by the initial prototype. Chapter 5: Development of a Novel Interface Pressure Transducer 96 Applied pressure (mmHg) Figure 5.3 Measured relationship between applied pressure and novel transducer output without equalizing pressure for test condition 1 i ' . " . r ' r_z. •. —x—test condition 1 —o—test condition 2 450 Applied Pressure (mmHg) Figure 5.4 Measured relationship between applied pressure and novel transducer output with equalizing pressure for test conditions 1 and 2 Chapter 5: Development of a Novel Interface Pressure Transducer 97 5.4.2 Second generation prototype transducer Several modifications were made to the initial prototype to attempt to fulfill some of the unmet optimal specifications. The transducer was modified to attempt to improve its sensitivity, radiotransparency, temperature sensitivity, and shear sensitivity. The sensitivity of a transducer is defined as the incremental output divided by incremental input. Therefore, the sensitivity of the strain-gauged diaphragm transducer without back pressure is defined as 1 mV (the smallest increment of voltage the voltmeter is capable of detecting) of output divided by the amount of input pressure required to obtain a change in the bridge output by 1 mV. The sensitivity of the initial prototype transducer did not meet the optimal specification for sensitivity defined as 1 mV / 2 mmHg. To increase the sensitivity, the diaphragm material was changed from a stainless steel shim stock to a mylar sheet 0.1 mm thick. Using this material, the sensitivity increased from approximately 1 mV / 25 mmHg to 1 mV / 2 mmHg which met the defined specifications. Use of mylar also made the transducer substantially more radiotransparent. While improved, the amount of remaining artifact produced by the small strain gauge on a mammographic x-ray image must be confirmed through testing the transducer in a mammography machine. The optimal specifications state that the transducer must also be insensitive to changes in temperature within the temperature range of the target applications. One benefit of a Wheatstone Chapter 5: Development of a Novel Interface Pressure Transducer 98 bridge configuration of resistors is that, with the proper resistor selection, the resistance changes due to ambient effects cancel out among first order terms as shown in the following equation: 5vo/v r e f = ( R 2 * 5 P M - R ! * 5 R 2 ) / ( R i + R 2 ) 2 - ( R 4 * 5 R 3 - R 3*5R4)/(R 3 + R 4 ) 2 If the ambient temperature is the same for all the resistors, temperature compensation can be achieved by ensuring all four resistors are identical (in value and material) so that their temperature coefficients are identical. Although, in the initial prototype, the strain gauge and three dummy resistors used were the same value, their temperature coefficients were different due to their different constructions. To improve the temperature compensation of the second generation prototype, four identical 120 Ohm strain gauges (Micro-measurements, Raleigh, North Carolina, USA) were used in the construction of the Wheatstone bridge circuit. One strain gauge was bonded to the diaphragm while the remaining three strain gauges made up the three passive arms of the bridge. One potential difficulty with using a strain gauge to monitor the position of the diaphragm is that shear strains transmitted by friction between the interface material and strain gauge or diaphragm could potentially effect the output of the strain gauge, yielding an inaccurate interface pressure measurement. The second generation prototype was constructed such that the strain gauge was bonded to the underside of the diaphragm. The strain gauge was therefore in contact with a static fluid which, by definition, cannot sustain substantial shear stress. Therefore shear strains could not develop in the strain gauge due to the fluid contact. However, the strain gauge Chapter 5: Development of a Novel Interface Pressure Transducer 99 could still be affected by shear strains caused by friction between the interface material and diaphragm. Also, it was necessary to distort the diaphragm seal to the chamber to feed the wire leads out from the strain gauge. The effects of all these modifications on transducer performance were then evaluated in the calibration system under a variety of test conditions. A further modification was the use of a shim (a material the same thickness as the transducer used to surround the transducer) which was evaluated for its effect on transducer output (Figure 5.5). The shim was constructed from modeling clay which was easy to modify in the event of a change in transducer design. modeling clay transducer embedded into modeling clay of the same thickness / Figure 5.5 Schematic of transducer and shim Chapter 5: Development of a Novel Interface Pressure Transducer 100 5.4.3 Results of second generation transducer testing The second generation prototype with the aforementioned modifications was evaluated in the calibration system using test conditions 1 and 2 (Section 4.3.1). Although the sensitivity of the transducer was greatly improved, the measured relationship between the applied pressure and the transducer output demonstrated increased dependence on the compliance of the interface materials (Figure 5.6). o— [test condition 2 \—x— test condition 2 (suffer gel) '450 T test condition 1 400 350 X S 300 CD V -3 250 C/5 P 200 150 Q -"3 100 O 50 0 0 50 100 350 400 450 150 200 250 300 \ Applied Pressure (mmHg) • Figure 5.6 Measured relationship between applied pressure and 2nd generation prototype output The change in diaphragm material from stainless steel shim stock to a mylar film may account for the greater scatter in transducer output for the different interface materials used during testing. Stainless steel shim stock has a high Young's modulus and behaves linear elastically within the pressure range it was tested. Mylar film has a much lower Young's modulus which contributed positively to the transducer performance by increasing its sensitivity. However, stresses in the diaphragm due to friction between the interface materials and the Chapter 5: Development of a Novel Interface Pressure Transducer 101 diaphragm may have translated into higher strains in the mylar than in the stainless steel, which may partially explain the higher level of transducer output scatter with the mylar diaphragm. A comparison of the initial prototype with the second generation prototype demonstrates both improvements and new difficulties (Table 5.1) Modifications to: Initial prototype Second generation prototype Diaphragm material Stainless steel diaphragm • . sensitivity of 1 mV / 25 mmHg • not radiotransparent Mylar diaphragm • sensitivity of 1 mV / 2 mmHg • diaphragm radiotransparent however further testing is required to determine the artifact left by the strain gauge on an x-ray image Wheatstone bridge circuit components Wheatstone bridge circuit constructed from a strain gauge and three dummy resistors of the same resistance but different temperature coefficients. • sensitive to changes in ambient temperature Wheatstone bridge circuit constructed from four identical strain gauges. • insensitive to changes in temperature (as long as temperature is the same for all resistors) Placement of the diaphragm strain gauge The strain gauge was bonded to the top surface of the diaphragm exposing it to stresses from interface material contact. • susceptible to shear stresses in the diaphragm The strain gauge was bonded to the under surface of the diaphragm which is in contact with fluid however the diaphragm may still experience shear stresses due to friction forces against the interface material. • less susceptible to shear stresses in the diaphragm Table 5.1 Comparison of initial and second generation prototypes Chapter 5: Development of a Novel Interface Pressure Transducer 5.4.4 Third generation prototype transducer 102 A third generation prototype was developed to address the unmet optimal specifications of the second generation prototype and to tailor the transducer for use in surgical retraction and tourniquet applications. The base used in the initial and second generation prototypes was not acceptable for use in surgical retraction or tourniquet applications due to its overall dimensions of 50 mm square x 6 mm thick. A thinner, rectangular base was designed with dimensions of 60 mm x 16 mm x 3 mm. The circular chamber was 8 mm in diameter and approximately 1.5 mm in depth. Aluminum was used as the base material due to its machinability and availability in 3 mm plate stock. Also, neither surgical retraction nor tourniquet applications require that the transducer be radio-transparent. Due to the thinner profile of the transducer base, a smaller 1.5 mm air hole and tubing were utilized. The lead wires in the second generation prototype passed between the diaphragm and chamber base were found to cause a disruption in the diaphragm bond. In order to avoid this, the lead wires were fed through a separate exit hole drilled into the chamber base in the third generation prototype. The lead wires were bent at a 90 degree angle approximately 1 mm from the strain gauge soldering tabs to exit the chamber. However, the gauge of the lead wires used in the initial and prototype transducers proved to be too stiff. The wires acted as a cantilever, preventing the diaphragm from moving freely in a vertical direction. Thinner, more flexible lead wires were required to reduce the constraint effects on the diaphragm response. Shellac coated, Chapter 5: Development of a Novel Interface Pressure Transducer 103 copper transformer wire was chosen as a more suitable lead wire and the transducer was constructed utilizing this flexible wire (approximately 0.1 mm in diameter). The bond between the diaphragm and the base was also improved in the third generation prototype transducer. Although the film transfer adhesive used in the initial and second prototype was easy to apply, it was not a suitable permanent adhesive. Two other adhesives were tested as possible alternatives. First, a cyanoacrylate adhesive, typically used to bond the strain gauge to the mylar diaphragm, was used to bond the diaphragm to the base with limited success. To attain a complete bond with cyanoacrylate requires a molecular layer thickness therefore pressure and lack of motion are essential during bonding. The diaphragm bonding was complicated by the lead wires which needed to be treated with care to ensure the soldering bonds didn't break during the process. As a result, the cyanoacrylate adhesive was not successful and a second adhesive was tested. Five minute setting two-part epoxy (Cole Parmer Instrument Co., Chicago, EL, USA) was chosen because it allowed time for application and careful placement of the diaphragm on the chamber base while ensuring the lead wires were not under tension. Once the proper placement of the diaphragm was achieved, finger pressure was applied while the epoxy set. In approximately five minutes the bonding reaction was completed and an air tight seal was achieved. Chapter 5: Development of a Novel Interface Pressure Transducer 104 5.4.5 Results of third generation transducer testing The measured relationship between applied pressure and the transducer output using test condition 1 demonstrated a 25 mmHg offset (Figure 5.7). Theory predicted an output pressure equivalent to the applied pressure. The offset represented strain in the diaphragm not related to the normal component of applied pressure. One possible factor was that strain was due to shear stresses which developed between the diaphragm and the interface material. Also, the output gradually decreased per unit of applied pressure producing a slight non-linearity in the output. This non-linearity also represented a strain in the diaphragm not related to the normal component of applied pressure. It was observed that pressure applied to the edges of the transducer base produced strain in the diaphragm due to the bending moment induced by the applied pressure. In an attempt to reduce the bending moment, the transducer base was shortened from 60 mm to 30 mm, while maintaining the diaphragm in the centre of the base. The transducer was re-tested in this configuration under test condition 1 and compared to the results from the 60 mm transducer (Figure 5.8). Comparing the results, the strain due to bending seemed to have been reduced, eliminating the non-linear output, however the offset was still present. This was to be expected as the shear stresses had not been eliminated. Finally the transducer was incorporated into the shim and evaluated using test conditions 1 and 2. The measured relationship between the applied pressure and the transducer output demonstrated a dependence on the compliance of the interface materials (Figure 5.9). It was theorized The compliance-dependent behaviour may be caused by bending moments or shear Chapter 5: Development of a Novel Interface Pressure Transducer 105 third generation reference line (1:1) Applied Pressure (mmHg) Figure 5.7 Measured relationship between applied pressure and the third generation prototype output . _ _ Applied Pressure (mmHg) Figure 5.8 Measured relationship between applied pressure and third generation prototypes using test condition 1 Chapter 5: Development of a Novel Interface Pressure Transducer 106 Applied Pressure (mmHg) Figure 5.9 Measured relationship between applied pressure and third generation prototype output forces which may be related to the interface material compliance. Furthermore, the behaviour may be a result of the stronger diaphragm bond which could transmit shear and bending stresses to the diaphragm. Several significant differences exist between the second and third generation prototypes and these differences may have contributed to the compliance-dependent behaviour exhibited by the third generation transducer. A summary of the modifications and resultant behaviours demonstrated the need for a fourth generation prototype to overcome these limitations (Table 5.2). Chapter 5: Development of a Novel Interface Pressure Transducer 107 Modifications to: Second generation prototype Third generation prototype Transducer base material Ultra-high molecular weight polyethylene • radiotransparent Aluminum • not radiotransparent Transducer base geometry 50 mm x 50 mm x 6 mm • not suitable for surgical retraction or tourniquet applications 3 0 mm x 15 mm x 3 mm • may experience bending moments and resulting stresses on the diaphragm may produce erroneous interface pressure readings • bending moments may be affected by the compliance of the interface materials possibly leading to compliance-dependent behaviour Diaphragm adhesive Film transfer adhesive • diaphragm bond not air tight Five minute setting two-part epoxy • air-tight diaphragm bond • may transfer stresses due to bending and shear possibly leading to interface material compliance-dependence Lead wires x gauge wire • may constrain diaphragm motion y gauge shellac coated transformer wires • does not constrain diaphragm motion in the vertical direction Table 5.2 Comparison of second and third generation prototypes 5.4.6 Fourth generation prototype transducer A fourth design was constructed to reduce the strains in the diaphragm caused by shear and bending moments in the transducer base. Following these modifications, the transducer was tested in the calibration system using test conditions 1 and 2. Chapter 5: Development of a Novel Interface Pressure Transducer 108 To reduce bending moments in the transducer base, the shape was changed from a 30 mm x 15 mm rectangle to a 24 mm diameter disk. Details of the new transducer design are shown in Figure 5.10. Changing the transducer base material from aluminum to steel would reduce the strains due to bending by approximately a factor of three for the equivalent amount of stress due to steel's higher modulus of elasticity. Two identical transducers were constructed following the circular design, one from aluminum and one from steel to test the effect due to the base material. '2.5 8dia -14dia -16dia 24dia 1.5 3.0 Figure 5.10 Schematic of the fourth generation transducer design It was hypothesized that shear stresses due to friction between the interface material and the diaphragm may have significantly effected the transducer output. To reduce any shear stresses present, a layer of liquid gel (Echo-gel, Lead-Lok Inc., Idaho, USA ) was placed between the transducer diaphragm and the interface material. The layer was less than 1 mm thick and was free to move laterally without any restriction. Another problem was that the applied Chapter 5: Development of a Novel Interface Pressure Transducer ' 109 \ pressures were causing the modeling clay shim to compress under the interface materials. Therefore, the shim material was changed from modeling clay to an aluminum to maintain the shim thickness. 5.4.7 Results of fourth generation transducer testing The measured relationships between the applied pressure and the transducer output for the third generation transducer and fourth generation aluminum and steel circular transducers with the clay shim subject to test condition 1 demonstrated that the circular design improved the transducer output significantly by reducing the offset (Figure 5.11). The output from the aluminum transducer indicates that there is still a small amount of bending stress, however the higher elastic modulus of the steel reduced the effect of this stress. The measured relationship between applied pressure and the output pressure for the steel transducer for conditions 1 and 2 with the applied liquid gel and aluminum shim demonstrated that the transducer's output was independent of interface material compliance (Figure 5.12). It appears that this combination of transducer design, transducer base material, gel, and shim Chapter 5: Development of a Novel Interface Pressure Transducer 110 -o-Aluminum base Mild steel base -x- third generation 0 50 100 150 200 250 300 350 400 450 Applied Pressure (mmHg) Figure 5.11 Measured relationship between the applied pressure and the fourth generation prototypes'outputs -o-Ztest condition 1 —x— 'test condition 2 — t e s t condition 2 (stiffer gel ""500 -r ~ " " ^  " ' " "' ~ " ' "~ ' Applied Pressure (mmHg) Figure 5.12 Measured relationship between the applied pressure and the fourth generation steel transducer prototype with liquid gel and aluminum shim Chapter 5: Development of a Novel Interface Pressure Transducer 111 material has effectively reduced the shear stresses and bending strains which previously significantly effected transducer output. A table of the modifications and their effects shown below documents the significant differences between the third and fourth generation transducer prototypes (Table 5.3). Modifications to: Third generation prototype Fourth generation prototype Base geometry 30 mm x 15 mm x 3 mm 24 mm diameter x 3 mm depth • reduced bending moments Base material Aluminum Steel • reduced strains due to bending and shear forces De-coupling agent None Ultrasonic gel • reduced shear stresses on the diaphragm Shim material Modeling clay • deformed under applied loading Aluminum • maintained its shape Table 5.3 Comparison of third and fourth generation prototypes 5.4.8 Comparison to optimal specifications The fourth generation interface pressure transducer has been modified to meet most of the optimal specifications defined in Section 2.5.2 under laboratory conditions. The following is a description of the transducer performance under calibration system test conditions. • the transducer measures pressure in a direction normal to the device/material interface Chapter 5: Development of a Novel Interface Pressure Transducer 112 • the area of sensing element less than 10 mm in diameter • the transducer measures pressure in the range of 0 - 500 mmHg and • the sensing element is capable of being incorporated into a linear (1 x3) or rectangular (3x3) array and measuring pressure simultaneously with the appropriate control system • the transducer output is independent of the compliance characteristics of the interface materials • using a steel base, the transducer is not suitable for imaging applications such as mammography, however, the second generation prototype which is constructed from a radiotransparent high density polyethylene would be a suitable transducer • the absolute error of the transducer is within the +/- 2 mmHg limit • the error due to hysteresis and/or drift over a one hour time period is within the +/- 2 mmHg limit • the frequency response of a strain gauge is sufficient to accommodate the target applications, however an appropriate control system needs to be designed to auto-regulate the diaphragm balancing with an appropriate frequency response for a given application • the modes of failure include breaking of leads which would be obvious to the operator and a disruption of the diaphragm seal which at this point in the transducer's development is not detectable unless the transducer is visually inspected • the transducer is compensated for changes in ambient temperature that effect the entire Wheatstone bridge circuit, however the transducer has not been tested for sensitivity to Chapter 5: Development of a Novel Interface Pressure Transducer 113 changes in temperature of the active strain gauge compared with the three remaining dummy gauges in the bridge circuit • the transducer can be calibrated and its calibration can be checked by a clinical user in application environments • the calibration-checking process is fast, convenient and intuitive • the transducer specifically avoids any potential electrical or thermal hazards • the transducer can be sterilized by either gamma radiation or ethylene oxide but will not withstand steam sterilization by autoclaving • the transducer could be developed further to meet the cost specifications for both reusable and disposable transducers 5.5 Summary One of the main goals of this research was to develop and evaluate an interface pressure transducer for use in biomedical applications. A transducer technology was selected for development based on the findings of the previous transducer evaluations (Chapter 4). Four generations of prototypes were developed before a satisfactory compliance-independent interface pressure transducer was realized. Some of the essential characteristics of the successful prototypes included the use of a liquid gel to reduce shear stresses and a circular, steel, transducer base to minimize strains due to bending and shear loads. The transducer met many of Chapter 5: Development of a Novel Interface Pressure Transducer 114 the optimal specifications under laboratory testing in the calibration system and was deemed ready for integration into medical devices for evaluation in the selected target applications (Chapter 6). 115 6 E V A L U A T I O N O F T H E D E V E L O P E D T R A N S D U C E R IN D E M O N S T R A T I O N APPLICATIONS 6.1 Chapter overview An important objective of this research was to evaluate the developed transducer in target applications. In this chapter, the novel interface pressure transducer was evaluated in two of the target applications, namely surgical retraction and tourniquets. The transducer was not evaluated in the mammography application due to a lack of accessibility to a mammography machine. Fortunately, this application has the least complicated measurement environment (i.e. a flat interface). Interface pressure measurement in surgical retraction is required to study the relationship between retractor pressure, application times, and tissue damage, as well as for the safe control of automated surgical retractors. The transducer was clinically evaluated in the surgical retractor application by integrating it into a surgical retractor blade and testing it in a series of tests involving pigs. The objectives of the tests were two-fold: to evaluate the developed transducer under in-vivo conditions; and to investigate the relationship between surgical retraction pressure, application time and abdominal tissue damage. The decision to conduct the tests using pigs was based on the availability of animals that were already undergoing a separate clinical trial. It was possible to conduct the retractor tests concurrently, thereby ensuring the optimal use of Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 116 resources. The tests involved making an incision in the abdomen and retracting the skin, fat and muscle tissues on both sides of the incision with a self-retaining retractor for several hours. The transducer output was monitored intermittently and the calibration of the transducer was checked at various intervals. The transducer was found to operate satisfactorily under test conditions. Tissue samples were taken at the end of the tests to analyze them for damage based on criteria established by a pathologist. The sampled tissue showed no signs of damage from the retraction indicating that longer retraction times and higher retraction pressures could have been sustained before tissue damage occurred. Interface pressure measurement can be employed to. improve the design, safety and operation of both pneumatic and non-pneumatic transducers. The developed transducer was further evaluated under both pneumatic and non-pneumatic tourniquet cuffs to determine its usefulness in this application. The transducer was tested first under the centre, and then under the edge of. the pneumatic cuff and it was found to respond well in the centre location. Due to perturbation of the interface, the transducer performance in the edge location was not satisfactory. Perturbation of the interface curvature was responsible for poor transducer ' performance under the non-pneumatic cuff. Further research and development is needed to minimize the perturbation of the interface in tourniquet applications. Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 111 6.2 Clinical evaluation of the novel transducer on a surgical retractor blade The relationship between retraction pressure and tissue damage is currently not well defined, however, several studies have demonstrated that tissue damage can be caused by retraction pressure (as described in the literature review). It was found that a higher retraction pressure applied for a longer time period causes more damage than a lower retraction pressure applied for a shorter time period. Tissue damage is of major concern in applications involving delicate organs, such as the brain or liver, because damage could cause a permanent debilitating injury. The measurement and control of retraction pressure is also important in the development of automated or robotic surgical retractors. If retraction pressure can be monitored in these procedures, it may be used to help ensure that pressures do not exceed a safe magnitude and application time. The developed interface pressure transducer was evaluated in the calibration system and demonstrated the ability to measure interface pressure under varying material compliance conditions. Preliminary in-vivo evaluations of the developed transducer were then undertaken as described below. Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 118 6.2.1 Objectives of surgical retractor trials The objectives of this preliminary surgical evaluation were two-fold: a) to test the novel interface pressure transducer under in-vivo conditions and b) to investigate the relationship between surgical retraction pressure, application time, and abdominal tissue damage. 6.2.2 Materials and Methods The developed transducer was prepared for use in the surgical trials. To integrate the interface pressure transducer into a surgical retractor, it was placed in an aluminum shim and bonded to the blade of a self-retaining chest retractor with five minute setting epoxy (Cole Parmer Instrument Co., Chicago, II, USA) (Figure 6.1). The instrumentation was set-up for transducer operation (Section 5.3.3). The transducer was covered with a thin layer of ultrasonic coupling gel. A layer of latex membrane material, approximately 1 mm thick, was placed over the gel and ensured that the gel remained over the transducer diaphragm throughout the tests. Figure 6.1 Transducer integrated into shim bonded to retractor blade Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 119 Female pigs undergoing a separate trial underwent the surgical retractor trial simultaneously. The test procedure was performed as follows. Each pig was first anesthetized. . A 45 mm incision was made in the abdomen in the proximal-distal direction, beginning 75 mm proximal to the pubic bone. The incision was made through the skin, fat and muscle tissues. The self-retaining retractor with integrated interface pressure transducer was used to retract the tissues (Figure 6.2). V- 1 1- ~ Figure 6.2 Retractor blades retracting tissue Displacement-controlled and pressure-controlled procedures were conducted to investigate the tissue behaviour. In the displacement-controlled procedure, the retractor blades were held 55 mm apart and the retraction pressure was monitored at regular intervals (approximately fifteen minutes). In the pressure-controlled procedure, the retractor blades were initially set 55 mm apart and subsequently opened further to maintain the interface pressure Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 120 recorded after the first five minutes of retraction time (the pressure was a function of the properties of the tissue which varied in each test). Calibration checks were made at various intervals to ensure the same input/output relationship existed throughout the trial and the transducer was performing properly. The calibration check involved unloading the transducer and checking that the transducer output dropped to zero. Since the animal was sacrificed at the end of the trial, sterilization of the surgical retractor with the integrated transducer was not performed. When the test animal was sacrificed, a sample of retracted tissue including the skin, fat, and muscle surrounding the incision was removed. The tissue sample was then preserved in formalin in preparation for analysis.. The presence of tissue damage was investigated by conducting a histological analysis of the tissue samples. Each tissue sample was divided into six zones which sustained different loading conditions (Figure 6.3). Thin sections from the middle of each retracted tissue, the edge of the retractors, and the ends of the incision were taken, prepared, and stained. They were analyzed by Dr. D A . Owen, Head, Anatomical Pathology, Vancouver Hospital and Health Sciences Centre, for tissue damage employing criteria including D N A leakage, vascular congestion and thrombi using an hemotoxylin and eosin stain. Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 121 Proximal x Left 2 3 T\ incision, 45 mm lone Right Distall _ • Figure 6.3 Tissue sample zones 6.2.3 Results and discussion The transducer operated successfully through five clinical trials and was not adversely affected by the measurement environment. To maintain proper transducer operation it was essential that the coating of ultrasonic coupling gel was maintained by the latex covering. Proper transducer operation was validated by calibration checks. Five trials provided relationships between the distance between the retractor blades, application time and retraction pressure. The relationship obtained between the application time and retraction pressure for a given distance between the retractor blades illustrated with visco-Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 122 elastic behaviour and was indicated in the three displacement-controlled trials (Figures 6.4 and 6.5). The two pressure-controlled trials indicated that the distance between retractors blades increased steadily with time to maintain a constant retraction pressure (Figure 6.6). These findings were expected, similar results have been reported in the literature [3]. However, the number of trials conducted did not yield enough data to develop a mathematical relationship which definitively relates the distance between the retractor blades, the application time, and the retraction pressure. Figure 6 . 4 First displacement-controlled trial retraction pressures Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 123 Time (minutes) Figure 6.5 Second and third displacement-controlled trial retraction pressures Figure 6.6 Retractor blade displacements for pressure-controlled trials Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 124 As the second objective was to investigate the relationship between surgical retraction pressure, application time and abdominal tissue damage, samples from two displacement-controlled trials were analyzed for signs of tissue damage. In one trial, the retractor had been applied for three hours without removal. In the second trial, the retractor had been applied for one hour without removal. In both trials, calibration-checking was not performed until the end of the trial because during this process the tissue would have been unloaded allowing a fresh supply of blood to flow to the tissues. This blood supply would have reduced the tissue damage. In both cases the tissue showed no signs of damage due to the retractor. The only sign of damage was the zone of tissue necrosis due to the cauterizing instrument used to open the incision. These results indicate that the level and duration of retractor pressure applied was not sufficient to cause damage in the abdominal tissue of a female pig. Further clinical trials are required to determine the complex relationship between retraction pressure, application time and tissue damage in humans. 6.2.4 Conclusions The surgical retractor trials demonstrated the use of the interface pressure transducer in a clinical application. They also helped provide a better understanding of the conditions riecessary for optimal performance of the transducer, as well as the behavior of tissue under retractor pressure over time. Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 125 The interface pressure transducer operated properly when a layer of ultrasonic coupling gel was applied to the diaphragm and a layer of latex was placed over the gel. The latex acted as a barrier so that the gel was not absorbed into the tissue or get wiped off the retractor blade during insertion into the incision. It was hypothesized that the gel reduced shear stresses between the diaphragm and the interface material and, as such, the transducer diaphragm only responded to the normal component of pressure. The transducer zero was checked periodically and found to remain at zero as long as the gel/latex layers were intact. The displacement controlled trials indicated that the abdominal tissues of the female pigs behave as visco-elastic materials. This tissue behaviour will have a bearing on the parameters used for the control of automated surgical retractors. For example, it would not be feasible to simply set an automated retractor at a constant pressure because the retractor displacement would unnecessarily increase slowly throughout the surgery. A more feasible method of control would be to set a displacement and a limiting pressure which the retractor pressure could not exceed and then to attempt to open the incision with the retractor blades to the desired position. It is important to note that retraction of different parts of the anatomy will result in different tissue behaviours. Therefore further retraction pressure studies are necessary to develop suitable automated retractor controls for other areas of the anatomy such as brain or liver tissue. Lastly, the lack of tissue damage under trial conditions suggests that higher levels of pressure and/or longer duration times than applied here may be necessary before any signs of damage might appear. However, the results of these studies may not be valid for humans due to Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 126 inter-species variations. As well they are not valid for other more delicate tissues such as brain tissue or liver tissue where the literature has reported cases of tissue damage at much lower levels of retractor pressure. Further studies are required to investigate the relationship between retractor pressure and duration, and tissue damage in human tissues, especially more delicate tissues such as brain or liver tissue. 6.3 Laboratory evaluation of the novel transducer under pneumatic and non-pneumatic tourniquet cuffs A preliminary evaluation of the novel interface pressure transducer was conducted under both pneumatic and non-pneumatic tourniquet cuffs to determine its usefulness in this target application. ( N 6.3.1 Transducer performance under pneumatic cuff To move the perturbation of the interface away from the sensing element, the developed transducer was integrated into an aluminum shim. The shim radius approximated that of a large thigh so that its presence did not change the curvature of the existing limb model utilized in testing. The transducer and shim were then secured to a limb model with masking tape. A layer of coupling gel was placed on the transducer diaphragm and a latex membrane was placed on top Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 127 of the gel as a protective barrier. A pneumatic tourniquet cuff, approximately 100 mm in width, was wrapped tightly around the limb model such that the transducer was located on the mid-line of the cuff and secured with Velcro straps. Interface pressure measurements were recorded at 50 mmHg intervals as the cuff was inflated from 0 - 450 mmHg, then deflated back to 0 mmHg (Figure 6.7). It is generally assumed that the internal cuff pressure is equivalent to the applied interface pressure at the centre of the cuff where it has complete contact with the underlying limb. The transducer test results indicate that the transducer output correlated very well with the applied pressure in the centre of the cuff. The transducer and shim were then placed at the edge of the cuff and the cuff was inflated to 50 mmHg (Figure 6.8). In this configuration, the transducer's output indicated a negative applied pressure of approximately 25 mmHg which did not correlate well to expected therefore no further testing was conducted. This result may have been due t o the shim shape which did not extend across the width of the cuff thereby perturbing the cuff/limb model interface and altering the stress pattern on the transducer. Further transducer and shim development is required to determine the limitations of the novel interface pressure transducer for use in pneumatic tourniquet applications. Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 128 Cufflnflation Pressure (mmHg) Figure 6.7 Measured relationship between the cuff inflation pressure and the transducer output shim and transducer limb model Figure 6.8 Cross-sectional view of the placement of the transducer and shim under the edge of the cuff Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 129 6.3.2 Transducer performance under non-pneumatic cuffs The developed transducer was evaluated under a non-pneumatic; cuff placed on an acrylic limb model having an approximate radius of 50 mm to simulate a medium sized upper arm. A latex Esmarch bandage was used as the non-pneumatic cuff. A layer of urethane gel was placed between the transducer and limb model to simulate tissue. A modeling clay shim was built up around the transducer to move the edge effects away from the transducer diaphragm. The transducer and shim were secured to the limb model with masking tape. The bandage was wrapped around the limb model such that three layers were applied directly over the transducer diaphragm to approximate the number required to stop blood flow. To aid in the evaluation of the developed transducer, the pneumatic transducer with electrical switches (as described in Section 4.3.2.5) was placed on top of the developed transducer under the Esmarch bandage such that the sensing elements of both transducers were in alignment thus providing a second measure of the interface pressure. A pressure measurement from the developed transducer was taken (ensuring the pneumatic transducer was not inflated) followed by a measurement from the pneumatic transducer. This procedure was repeated three times (Table 6.1). Developed transducer Pneumatic electrical-switch transducer 54 mmHg 96 mmHg 56 mmHg 116 mmHg 42 mmHg 104 mmHg Table 6.1 Interface pressure measurements under an Esmarch bandage Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 130 The results indicate that the pneumatic electrical switch transducer consistently measured significantly higher pressures than those obtained with the developed transducer. It appears that the reason for the discrepancy was due to the flat surface of the developed transducer's diaphragm. A flat spot on the cylinder, such as that caused by the transducer diaphragm, may have caused a decrease in the radial component of force applied by the bandage over that local area (Figure 6.9). This force may have been shed to the rest of the bandage. In its current configuration the developed transducer is not suitable for use in non-pneumatic tourniquets due to the flat diaphragm. Since the radial component of force is the component of force we are trying to measure, it would be necessary to reconstruct the radius of curvature above the transducer diaphragm so that the bandage could maintain its cylindrical v shape and hence its radial component of applied force. (Figure 6.10). Further investigation is necessary to determine the feasibility of this approach. / / \ \ . \ Figure 6.9 Effect on applied pressure distribution due to a flat spot (not drawn to scale) Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 131 Figure 6.10 Pressure distribution with curved section restored (not drawn to scale) 6.4 Summary An important objective of this research was to evaluate the developed transducer in target applications. Due to ease of accessibility, surgical retraction and tourniquet applications were selected as demonstration applications. Although the transducer was not evaluated in the mammography application, it has the least complex measurement environment. It is reasonable to assume that, if the transducer works in more complex cases, it would be worth while pursuing the development of the transducer for use in mammography. Chapter 6: Further Evaluation of the Novel Interface Pressure Transducer in Demonstration Applications 132 It was found that the developed transducer operated properly in the surgical retraction trials conducted and seems to be suitable for this application. An essential component for its proper operation is the presence of a layer of coupling gel over the diaphragm and latex covering to protect the gel. The trials conducted demonstrated a lack of tissue damage under test conditions. Further studies are required to develop the relationships between retraction pressure, application time and tissue damage. In the tourniquet application, the transducer did not operate satisfactorily and further research and development is required to overcome the perturbation of the interface and pressure distribution due to the presence of the transducer. Based on the results of the transducer operation under surgical retraction, further development and evaluation of the transducer for mammography is recommended. 133 7 CONCLUSIONS AND RECOMMENDATIONS The final phase of this research work was to develop general conclusions about the need for interface pressure measurement and the development and evaluation of an interface pressure transducer for biomedical applications. As well, a summary of the contributions of this research to this field of study are outlined. Finally, recommendations for future work that wil l aid in the progression of the development and evaluation of an interface pressure transducer for biomedical applications are presented. 7.1 General conclusions The following conclusions were elicited from each chapter of this thesis: Chapter 2-Background and review of previous research 1) The need for measurement of the interface pressure between the respective biomedical devices and human anatomy to improve the efficacy and safety of the selected target applications of surgical retraction, surgical tourniquets and mammography was demonstrated. 2) The interface pressure transducers used in the selected target applications had limitations and none were suitable for use in all three applications. Chapter 7: Conclusions and Recommendations 134 3) The interface pressure transducers used in other biomedical applications reviewed were evaluated in a single application. 4) In most cases, the transducers were not calibrated under interface conditions. None of the transducers were evaluated or calibrated under interface materials of varying compliance. 5) In some cases, curvature of the interface of the device and tissue adversely effected transducer function. 6) In some cases, the perturbation of the interface of the device and tissue by the presence of the transducer adversely affected transducer function. 7) None of the transducers reviewed were suitable as an interface pressure transducer that was independent of interface material compliance. 8) The design criteria and optimal design specifications for a compliance-independent, interface pressure transducer were developed. 9) The use of a calibration technique that simulated interface, conditions when calibrating an interface pressure transducer was essential to ensure that the transducer was not adversely effected by interface materials of varying compliance. Chapter 3-Finite element model of measurement environment and transducer 1 ) Shear stresses were present at the interface of the tissue in contact with both the device and transducer in both the flush and protruding transducer models. 2) The interface loading may be significantly different that the loading applied to the device. Chapter 7: Conclusions and Recommendations 135 Chapter 4-Development of a calibration system and evaluation of interface pressure transducers 1) None of the transducers tested were satisfactory as interface pressure transducers in all three target applications in their existing configuration and packaging. Limitations of these transducers included unacceptable levels of hysteresis and drift, radio-opacity, and dependence on the interface material compliance. 2) Transducers which related applied pressure to the deflection of their sensing element were highly dependent on the compliance of the interface materials. 3) In theory, if a method of measuring the applied pressure could be achieved without deflection of the sensing element, dependence on the interface material compliance could be minimized or eliminated. As such, further development of a multi-application interface pressure transducer involved the development of this essential characteristic. Chapter 5-Development of a novel interface pressure transducer 1) An effective interface pressure transducer was developed which operated by balancing interface pressure on a diaphragm with an equal and opposite fluid pressure The fluid pressure is theoretically equivalent to the average applied interface pressure if the diaphragm is maintained in its original, unloaded position. A strain gauge was bonded to the diaphragm and connected to a Wheatstone bridge circuit to monitor the position of the diaphragm. Chapter 7: Conclusions and Recommendations 136 2) The following characteristics were necessary to ensure that the transducer met the key optimal specifications previous defined: • diaphragm material chosen was mylar to give the required transducer sensitivity • the resistors in the Wheatstone bridge circuit were identical strain gauges to improve the temperature sensitivity • the strain gauge was bonded to the underside of the diaphragm to reduce the shear stresses due to the interface material • shellac coated, copper transformer wire was chosen as the lead wires from the strain gauge on the diaphragm to reduce the diaphragm loading • five-minute setting epoxy was used to bond the diaphragm to the base to get a air-tight bond • the body of the transducer was chosen to be circular to minimize any strains due to bending moments • steel was chosen as the material for the body of the transducer in the surgical retraction application to minimize any strains due to bending moments • an aluminum shim was added to the transducer to ensure the transducer did not perturb the interface pressure distribution • a thin layer of ultra-sound liquid gel was placed between the transducer diaphragm and the i interface material to reduce the interface friction thereby reducing shear stresses 3) The transducer met many of the optimal specifications as described in Section 5.4.5. Chapter 7: Conclusions and Recommendations Chapter 6-Evaluation of developed transducer in demonstration applications 137 1) The developed interface pressure transducer operated optimally in the surgical retraction application when a layer of ultra-sonic gel was applied to the diaphragm and a layer of latex was applied to the gel. The latex acted as a barrier to protect the ultra-sonic gel from being absorbed into the tissues. 2) The surgical retraction trials indicated that the abdominal tissues of the female pig behave as visco-elastic materials which may be represented by Maxwell's model. As such, the tissue behaviour has a bearing on the parameters used for the control of automated surgical retractors. 3) The lack of tissue damage under the surgical retraction trial conditions indicates that higher levels of pressure and/or longer duration times were necessary before any signs of damage would appear in the pig tissue. 4) The developed transducer worked well under the centre of a pneumatic tourniquet cuff. 5) The evaluation of the developed transducer under a non-pneumatic cuff demonstrated the potential for the perturbation of the interface to adversely affect the transducer output. Chapter 7: Conclusions and Recommendations 138 7.2 Contributions of the research The principal contributions of the work described in this thesis were as follows: 1) Optimal specifications for a generic biomedical interface pressure transducer were developed for a broad range of existing and future applications in biomedical engineering, i.e. measurement of pressure at the interface between medical devices and living tissues, organs, and limbs; 2) Models were developed for both the measurement environment and transducer to gain a better understanding of the mechanical response of the environment/transducer system, and to be instrumental in the design of improved interface pressure transducers; 3) A calibration system was developed to simulate interface conditions and allow for the testing of transducers, including commercial and prototype transducers, under a range of simulated tissue compliances and shapes, and to permit testing of any newly developed transducers or different interface environments; 4) Using the calibration system, testing of a broad range of commercial and prototype transducers based on a comprehensive review of the literature revealed that none of these transducers met all of the key specifications; 5) A novel balanced-diaphragm transducer was developed in an effort to meet the optimal specifications for an interface pressure transducer; 6) This novel balanced-diaphragm interface pressure transducer was evaluated in the calibration system and was determined to meet the optimal performance specifications in laboratory testing; and Chapter 7: Conclusions and Recommendations 139 7) The balanced diaphragm interface pressure transducer was further evaluated in a demonstration application, surgical retraction, and was found to meet optimal performance specifications under clinical conditions. 7.3 Recommendations for future work The recommendations for future work have been separated into three categories, namely, the finite element model, the novel transducer development, and the demonstration and target applications. 1) Finite element model • determine the effect of the interface material compliance and the transducer dimensions on the applied pressure distribution using the current model • modify the current model such that the effect of changing the interface profile (i.e. adding curvature) can be analyzed • modify the current model such that the tissue is modeled more closely to its actual properties (i.e. visco-elastic properties) • modify the current model such that the surface nodes of each component(i.e. device, transducer, and tissue) are not interconnected and can slide past one another ensuring that the friction is also modeled Chapter 7: Conclusions and Recommendations 140 • verify the results of the models with physical testing in the calibration system 2) Novel transducer development All applications • integrate a method of maintaining a thin uniform layer of ultra-sonic gel over the transducer diaphragm ensuring that the gel does not develop an internal pressure • develop an auto-regulation control system • develop a multi-sensor array of the novel transducer tailored to each application to measure . the pressure distributions between the devices and respective anatomy Mammography • test the artifact produced by the strain gauge on the mammogram Non-pneumatic tourniquet cuff • develop a method of maintaining the profile of the elastic bandage over the sensing element 3) Demonstration and target applications • conduct further surgical retractor trials to determine the relationship between retraction time and pressure versus tissue damage • conduct trials on delicate tissues which are at the highest risk of injury (i.e. brain, liver) to determine the relationship between retraction time and pressure versus tissue damage Chapter 7: Conclusions and Recommendations 141 • based on the results of these trials develop a set of guidelines for the safe use of hand-held, self-retaining, and automated surgical retractors • integrate the novel transducer array into a non-pneumatic tourniquet cuff to develop a low cost tourniquet system • integrate the novel transducer array into a mammography machine and run clinical trials to determine the current range of pressure distribution experienced by patients • based on the results of these trials develop a set of guidelines for the safe use of the mammography machine with the use of the novel transducer array 142 R E F E R E N C E S [I] D.J. Clark, IR. Chambers, K. Faulkner, J. Rayson, P.M. Hacking, and J. Milton, "Pressure measurements during automatic breast compression in mammography", Journal of Biomedical Engineering, vol. 12, September, 1990. [2] J. Findlay, "Development and evaluation of an automated pre-robotic system and an advanced robot for surgical retraction", The University of British Columbia, Department of Electrical Engineering, M.A.Sc. thesis, November 1990. [3] Y. Kawaguchi, H. Matsui, and H. Tsuji, "Back muscle injury after posterior lumbar spine surgery", Spine, vol. 19, no. 22, pp. 2590-2597, 1994. [4] D. Aserman, "Controlled hypotension in neurosurgery with hexamethonium and procaine amide", British Medical Journal, vol. 1, pp. 961-964, 1953. [5] F.M. Ameli et. al., "Safety of cholecystectomy with abdominal aortic surgery", Canadian Journal of Surgery, vol. 30, no. 3, pp. 170-173, 1987. [6] H. Lozman et. al., "Thrombosis of peripheral aneurysms. A complication of colorectal surgery", Diseases of the Colon and Rectum, vol. 26, no. 3, pp. 167-169, 1983. [7] N.K. Schoondorf, "On the prevention of paresis of the femoral nerve following abdominal gynaecological operations", Geburtshilfe UndFrauenheilkunde, vol. 42, no. 1, pp. 58-62, 1982. [8] R.K. Laha et. al., "Cerebellar retraction: significance and sequelae", Surg. Neurol., vol. 12, pp. 209-215, 1979. [9] M . Albin et. al., "Clinical and experimental brain retraction pressure monitoring", Acta. Neurol. Scand, vol. 56 (suppl. 64), pp.522-223, 1977. [10] A. Yokoh et. al., "Intermittent versus continuous brain retraction", J. Neurosurg., vol.58, pp. 918-923, 1983. [II] K. Hongo et. al., "Monitoring retraction pressure on the brain", J. Neurosurg., vol.66, pp. 270-275, 1987. [12] J.A. McEwen, Pneumatic tourniquet, U.S. Patent No. 4,469,099, September 4, 1984. [13] J. A. McEwen and R.W. McGraw, "An Adaptive tourniquet for improved safety in surgery", IEEE Transactions on Bio-Medical Engineering, BME-29, pp. 122-129, 1982. ? 143 [14] T.L. Johnson, S.C. Wright, arid A. Segall, "Filtering of muscle artifact from the electroencephalogram", IEEE Transactions on Bio-Medical Engineering, BME-26, pp. 556-563,1975. [15] J.A. Shaw and D.G. Murray, "The relationship between tourniquet pressure and underlying soft-tissue pressure in the thigh", The Journal of Bone and Joint Surgery, vol. 64-A, pp. 1148-1152, 1982. [16] M . J. Bruner, "Time, pressure, and temperature factors in the safe use of the tourniquet", The Hand, vol. 2, pp. 39-42, 1970. [17] J.C. Griffiths and O.B. Heywood, "Bio-mechanical aspects of the tourniquet", The Hand, vol.5, pp.113-118, 1973. [18] M R . Moore, S.R. Garfin, and A.R. Hargens, "Wide tourniquets eliminate blood flow at low inflation pressures", The Journal of Hand Surgery, vol. 12A, pp. 1006-1011, 1987. [19] T.J. Fowler, G. Danta, and R.W. Gilliatt, "Recovery of nerve conduction after a pneumatic tourniquet: observations on the hind limb of the baboon", Journal of Neurology, Neurosurgery, and Psychiatry, vol. 35, pp. 638-647, 1972. [20] L. Klenerman, "Tourniquet time - how long?",. The Hand, vol. 12, pp. 231-234, 1980. [21] S. Patterson and L . Klenerman, "The effect of pneumatic tourniquets on the ultrastructure of skeletal muscle", The Journal of Bone and Joint Surgery, vol. 61-B, pp. 178-183, 1979. [22] W.P. Sy, "Ulnar nerve palsy possibly related to use of automatically cycled blood pressure cuff', Anaesthesia and Analgesia, vol. 60, pp. 687-688, 1981. [23] N. R. Fahmy and D.G. Patel, "Hemostatic changes and post-operative deep-vein thrombosis associated with use of a pneumatic tourniquet", The Journal of Bone and Joint Surgery, vol. 63A, pp. 461-465, 1981. [24] A.A. Hofmann and R.W.B. Wyatt, "Fatal pulmonary embolism following tourniquet inflation", The Journal of Bone and Joint Surgery, vol. 67-E;pp. 633-634, 1985. [25] R.E. Stein, "Use of the tourniquet during surgury in patients with sickle cell hemoglobinapothies", Clinical Orthopaedics, vol. 151, pp. 231-233, 1980. [26] J. Arenson, M.D. Steer, and N. Joffe, "Edema of the pipilla of water stimulating retained common duct stone", Surgery, vol. 817, pp. 712-713, 1980. [27] R.D. Kaufman, "Tourniquet induced hypertension", British Journal of Anesthesiology, vol. 54, pp. 333-336, 1982. 144 [28] J. C. Griffiths and M . Sankarankutty, "Bone marrow pressure changes under an inflatable tourniquet", The Hand, vol. 15, pp. 3-8, 1983. [29] R.W. Gilliatt, W.I. McDonald, and P. Rudge, "The site of conduction block in peripheral nerves compressed by a pneumatic tourniquet", Journal of Physiology, vol. 238, pp. 31p-32p, 1974. [30] F.H. Bentley and W. Schlapp, "The effects of pressure on conduction in peripheral nerve", Journal of Physiology, vol. 102, pp. 72-82, 1943. [31] J.A. McEwen, "Complications of and improvements in pneumatic tourniquets used in surgery", Medical Instrumentation, vol. 15, pp. 253-257, 1981. [32] A.C. McLaren and C.H. Rorabeck, "The pressure distribution under tourniquets", The Journal of Bone and Joint Surgery, vol. 67-A, pp. 433-438, 1985. [33] M . Breault, "A biomechanical investigation of blood flow occlusion acheived with the use of surgical pneumatic tourniquets", The University of British Columbia, Department of Mechanical Engineering, M.A.Sc. thesis, October 1988. [34] "Questions and Answers about Screening Mammography", Screening Mammogrpahy Program of British Columbia. [35] L. Tabar, G. Faberberg, N.E. Day, and L. Holmberg, "What is the optimum interval between mammographic screening examinations? An anlysis based on the latest results of the Swedish two-country breast screening trial", Br. J. Cancer, vol. 55, pp. 547-551, 1989. [36] K. Faulkner, I.R. Chambers, P. Hedley, and S.R. Thompson, "An instrument for monitoring the force applied by breast compression devices on mammography x-ray sets", Clin. Pys. Physiol. Meas., vol. 10, no. 2, pp. 177-180, 1989. [37] K. Faulkner, K.J. Robson, and C L . Kotre, "X-ray mammography and breast compression", The Lancet, pp. 797, September, 1992. [38] V.F. Andolina, S.L. Lille, and K . M . Willison, "Mammogrpahic Imaging: A practical guide", J. B. Lippincott Company, Philadelphia, 1992. [39] D.J. Watmough, and K . M . Quan, "X-ray mammography and breast compression"; The Lancet, pp. 122, July 1992. [40] T.A. Krouskop and S.L. Garber, "Interface pressure confusion", Decubitus, vol. 3, no. 3, August 1989. 145 [41] P.M. Burman, "Using Pressure Measurements to Evaluate Different Technologies", Decubitus, vol. 6, no. 3, May 1993. [42] M . Clark and L.B. Rowland, "Comparison of contact pressures measured at the sacrum of young and elderly subjects", Journal of Biomedical Engineering, vol. 11, May 1989. [43] A.E. Hover and T.A. Krouskop, "Pressure relief characteristics of a new foam overlay: A preliminary performance evaluation", Journal ET Nursing, vol. 19, no. 2, March/April 1992. [44] L.A. Petrie and R.S. Hummel, "A study of interface pressure for pressure reduction and reliefmattresses", Journal of Enterostomal Therapy, vol. 17, no. 5, September/October 1990. [45] L. Cron and S. Sprigle, "Clinical evaluation of the hemi-wheelchair cushion", The American Journal of Occupational Therapy, vol. 47, no. 2, February 1993. [46] S.A. Riedel, G.F. Harris, J.J. DeRosia, M.F. Pesigan, and H A . Jizzine, "An instrumented chair for assessing seated stability", Journal of Clinical Engineering, vol. 15, no. 6, Nov./Dec. 1990. [47] J.Y. Thompson-Bishop and C M . Mottola, "Tissue interface pressure and estimated subcutaneous pressures of 11 different pressure-reducing support surfaces", Decubitus, vol. 5, no. 2, March 1992 (accompanying letters). [48] K . M . Eckrich, "A pneumatic bladder array for measuring dynamic interface pressure between seated users and their wheelchairs", copywrite ISA, 1991 - Paper #91-018.0067-8856/91/135-140/$0+.50pp. [49] A. Patel, M . Kothari, J.G. Webster, W. Tompkins, and J. Wertsch, "A capacitance pressure sensor using a phase-locked loop", J. of Rehabilitation R &D, vol. 26, no. 2, pp. 55-62, 1989. [50] D.A. Hobson, "Comparative effects of posture on pressure and shear at the body-seat interface", Journal of Rehabilitation R &D, vol. 29, no. 4, pp. 21-31, 1992. [51] F A . Appoldt, L . Bennett, and R. Contini, "Socket pressure as a function of pressure transducer protrusion", Bulletin of Prosthetics Research, pp. 236-249, Spring 1969. [52] J.W. Rae and J.L. Cockrell, "Interface pressure and stress distribution in prosthetic fitting", Bulletin of Prosthetics Research, Fall 1971. [53] T. Van Pijkeren, M . Naeff, and H. Hok Kwee, "A new method for the measurement of normal pressure between amputation residual limb and socket", Bulletin of Prosthetics Research, vol. 17, no. 1, pp. 31-34, Spring 1980. 146 [54] H. Zhu, G.F. Harris, J.J Wertsch, W.J. Tompkins, and J.G. Webster, "A microprocessor-based data-acquisition system for measuring plantar pressures from ambulatory subjects", IEEE Transactions on Biomedical Engineering, vol. 38, no. 7, July 1991. [55] E . M . Hennig, PR. Cavanagh, H.T. Albert, and N H . Macmillan, "A piezoelectric method of measuring the vertical contact stress beneath the human foot", Journal of Biomedical Engineering, vol. 4, July 1982. [56] M . Lord, D. Reynolds, and J. Hughes, "Foot pressure measurement: a review of clinical findings", Journal of Biomedical Engineering, vol. 8, October 1986. [57] J.K. Roy, "Force, pressure, and motion measurements in the foot: Current concepts", Clinics in Podiatric Medicine and Surgery, vol. 5, no. 3, July 1988. [58] C A . Harries and S.P. Pegg, "Measuring pressure under burns pressure garments using the Oxford Pressure Monitor", Burns, vol. 15, no. 3, pp. 187-189, 1989. [59] J .C Barbenel and S. Sockalingham, "Device for measuring soft tissue interface pressures", Journal of Biomedical Engineering, vol. 12, November 1990. [60] P.M. Ferguson-Pell, "Design criteria for the measurement of pressure at body/support interfaces"', Engineering in Medicine, vol. 9, no. 4, 1980. [61] C L . Guthrie, "The effect of transducer curvature on interface pressure readings", RESNA 14th Annual Conference Proceedings, Kansas City, Mo. 1991. [62] K.H. Huebner and E.A. Thornton, "The Finite Element Method for Engineers", John Wiley and Sons, New York, 1982. [63] G.J. DeSalvo, J.A. Swason, "ANSYS User's Manual", Swanson Analysis Systems, Inc., June 1, 1993. [64] S.H. Advani, R.B. Martin, and W.R. Powell, "Mechanical properties and constitutive equations of anatomical materials", Applied Physiological Mechanics, pp. 31-103, Hardwood Academic Publishers, New York, 1979. [65] W.W. Chow and E.I. Odell, "Deformations and stresses in soft body tissues of a sitting person", Journal of BiomechanicalEngineering, vol. 100, pp. 79-87, 1978. [66] H. Yamada, "Strength of Biological Materials", Evans, Williams & Wilkins Co., Baltimore, 1970. 147 [67] A.J. Hodgson, "A Parametric study of the axisymmetric loading of an anisotropic elastic limb model by a tourniquet", University of British Columbia, Department of Mechanical Engineering, M.A.Sc. thesis, 1986. [68] S.M. Auerbach, " Axisymmetric finite element analysis of tourniquet application on limb", Journal of Biomechanics, vol. 17, pp. 861-866, 1984. / [69] Uniforce™, Force Imaging Technologies, Technical Note #101 (2/94) [70] Interlink Electronics, P.O. Box 40760, Santa Barbara, CA, USA. Product literature. [71] Pressure Monitors, Intracranial: Product Comparison System ECRI, April 1993. [72] Patent abstract from Steritek re: Human body presssure measuring appts.has constant gas . flow into plenum and maintains pressure equilibrium using diaphragm to control opening of exhaust tube. US Patent # 4393878. 148 APPENDIX A: Finite element model verification and program code i Appendix: Finite element program and model verification 149 Model verification The model was verified by: 1) selecting a Young's modulus (E) of 15000 N/m 2 and a Poison's ratio (u) of 0.3 for all the materials; 2) applying a negative vertical load (P) of 1500 N/m 2 to the device surface; 3) calculating the displacement and stress solution analytically; 4) obtaining the displacement and stress solution through the finite element program; and 5) comparing the analytical and finite element program solutions The analytical solution for the stress in the y-direction is: oy = P/A oy =-1500 N/m 2 Figure 3.3 shows the stress results of the finite element solution. This demonstrates that the finite element solution is equivalent to the exact analytical solution. axis of symmetry ANSYS 5.0 JUL 26 1994 15:08:20 PLOT NO. 2 NODAL SOLUTION STEP=1 SUB =1 TIME=1 SY (AVG) RSYS=0 DMX =0.001998 SMN =-1500 SMNB=-1500 SMX =-1500 SMXB=-1500 ^ _ -1500 -1500 N/m 2 Figure A . l F E M verification stress results Appendix; Finite element program and model verification 150 The analytical displacement solutions is as follows: Strain in the y-direction: sy = ay/E sy =-1500/15 000 , sy = -.1 For nodes on the top of the device, the distance from the constrained bottom edge is ty = 0.017 m The amount of displacment in the y-direction of these nodes due to the applied loading is: ey = ty * ey ey = 0.017 * -.1 ey =-0.0017 For nodes on the outside edge of the model, the distance from the axis of symmetry is &=0.035 m The amount of displacment in the x-direction of these nodes due to the applied loading is: ex = ex. * sx substituting for sx = - u * sy ex = & * -u *sy , ex = (0.035)*(-0.3)*(-.l) ex = 0.00105 The node that is common to the top and outside edge will see the largest displacement Dmax. Dmax = ( ey2 + ex2)1 / 2 Dmax = ((-0.0017)2 + (0.00105)2)1/2 Dmax = 0.001998 Appendix: Finite element program and model verification 151 Figure 3 . 4 shows the finite element displacement solution. D M X is the maximum displacement of the node that is common to both the top and outside edge and is equivalent to the analytically calculated maximum displacement. axis of asymmetry ANSYS 5.0 JUL 26 1994 15:07:49 PLOT NO. 1 DISPLACEMENT STEP=1 SUB =1 TIME=1 RSYS=0 DMX =0.001998 SEPC=0.626E-11 »DSCA=1 ZV =1 DIST=0.019828 X F Y F 018025 W 00765 CENTROID HIDDEN Figure A.2 F E M verification stress results The comparison of analytical solutions and finite element solutions shows that they are equivalent, therefore verifying that the model is functioning as required. Appendix: Finite element program and model verification 152 ANSYS P R O G R A M LISTING ********************** C*** C*** This program constructs the transducer and interface material models C*** /FILNAM.PRES2.DB /TITLE, Interface pressure transducer model /UNITS,SI * SI units throughout /PREP7 ETYPE STAT * Choose a staic analysis ET,1 ,PLANE2,,,1 * Choose element type ET,2,PLANE42,,,1 MP,EX, 1,150000000000 * Set material properties MP,NUXY,1,0.3 MP,EX,2,150000000000 MP,NUXY,2,0.3 MP,EX,3,15000 MP,NUXY,3,0.3 > MP,EX,4,15000 MP,NUXY,4,0.3 C*** c*** C*** Set up nodes and elements C*** C*** N,1,0,0 * Place nodes on model NGEN,5,1,1,1,1,0,2 NGEN,11,6,1,6,1,1,0 NGEN.12,,67,72,1,2,0 NGEN.6,1,1,1,1,0,2 NGEN,12,6,1,6,1,1,0 NGEN,13,6,67,72,1,2,0 N,145,0,11 NGEN.21,1,145,145,1, .5,0 NGEN,9,21,145,165,1,0,.5 N,334,11,11 NGEN.7,1,334,334,1,0,1 NGEN,13,7,334,340,1,2,0 N,425,0,15.5 NGEN,11,1,425,425,1,.5,0 N,436,0,16 NGEN,11,1,436,436,1,.5,0 N,447,0,17 NGEN,11,1,447,447,1,1,0 N,458,5.5,15.5 NGEN,10,1,458,458,1,.5,0 N,468,5.5,16 NGEN,10,1,468,468,1,.5,0 N,500,2.5,13 C*** C*** Appendix: Finite element program and model verification MAT,4 * Select material TYPE,2 * Select element type E,1,7,8,2 * Place elements on model EGEN,5,1,1 EGEN,11,6,1,5,1 E,67,73,74,68 EGEN,5,1,56,,1 EGEN,12,6,56,60,1 E,376,383,384,377 EGEN,4,1,116,,1 EGEN,12,7,116,119,1 E,145,146,167,166 EGEN,20,1,185„1 EGEN,8,21,185,204,1 TYPE.1 * Select element type E,6,146,145 E,6,12,146 E,12,147,146 E,12,148,147 E,12,18,148 E18,149,148 E,18,150,149 E,18,24,150 E,24,151,150 E,24,152,151 E,24,30,152 E,30,153,152 E,30,154,153 E,30,36,154 E,36,155,154 E,36,156,155 E,36,42,156 E,42,157,156 E,42,158,157 E,42,48,158 E,48,159,158 E,48,160,159 E,48,54,160 E.54,161,160 E,54,162,161 E, 54,60,162 E,60,163,162 E,60,164,163 E,60,66,164 E,66,165,164 E,165,376,186 E,376,377,186 E, 377,207,186 E,207,377,228 E,228,377,378 E,378,249,228 E,249,378,270 E,270,378,379 E,270,379,291 Appendix: Finite element program and model verification E.291,379,312 E,312,379,380 E, 312,380,333 TYPE,2 E,66,72,376,165 E,72,78,383,376 E,78,84,390,383 E,84,90,397,390 E,90,96,404,397 E,96,102,411,404 E,102,108,418,411 E,108,114,425,418 E,114,120,432,425 E,120,126,439,432 E.126,132,446,439 E,132,138,453,446 E,138,144,460,453 MAT, 2 * or MAT, 3 E,313,314,335,334 EGEN,10,1,400,,1 EGEN,2,21,400,409,1 MAT.1 TYPE.1 E,335,336,467 E,467,468,356 E,468,357,356 E,357,358,468 E,468,469,358 E,469,359,358 E,469,360,359 E,469,470,360 E,470,361,360 E,470,362,361 E,470,471,362 E.471,363,362 E.471,364,363 E.471,472,364 E,472,365,364 E,365,366,472 E,472,473,366 E.473,367,366 E,473,368,367 E,473,474,368 E.474,369,368 E,474,370,369 E,474,475,370 E.474,371,370 E.475,372,371 E,475,476,372 E,476,373,372 E,476,374,373 E,476,477,374 * Select element type * Select material type for protruding transd * Select material type for flush transducer * Select material type Appendix: Finite element program and model verification 155 E,477,375,374 TYPE,2 E,375,381,382,477 E,381,388,389,382 EGEN,12,7,451,,1 MAT, 3 E,323,324,345,344 EGEN,10,1,463 EGEN,2,21,463,472,1 TYPE,1 E,333,380,354 E,354,380,381 E,354,381,375 TYPE,2 E, 380,387,388,381 EGEN,12,7,486„1 C*** Set the boundary conditions NSEL,S,P50X,_ 19 1 2 3 4 5 6 145 166 187 208 . 229 250 271 292 313 334 355 467 DSYM,SYMM,X D,P50X,UY 1 D,P50X,UY 7 D,P50X,UY 13 D,P50X,UY 19 D,P50X,UY 25 D,P50X,UY 31 * Select element type * Select material type * Select element type * Select element type * Place axis of symmetry * Constrain nodes in the y-direction Appendix: Finite element program and model verification 156 D,P50X,UY 37 D,P50X,UY 43 D,P50X,UY 49 D,P50X,UY 55 D,P50X,UY 61 D,P50X,UY 67 D,P50X,UY 73 D,P50X,UY 79 D,P50X,UY 85 D,P50X,UY 91 D,P50X,UY 97 D,P50X,UY 103 D,P50X,UY 109 D,P50X,UY 115 D,P50X,UY 121 . D,P50X,UY 127 D,P50X,UY 133 D,P50X,UY 139 C*** c*** C*** Set the pressure distribution C*** C*** NSEL,S,P50X_ 24 467 468 469 470 471 472 473 474 475 476 477 Appendix: Finite element program and model verification 157 382 389 396 403 410 417 424 431 438 445 452 459 466 SF,ALL,PRES,1500 SAVE,PRES2,DB FINISH C*** C* c*** C*** This program executes the program C*** /SOLU SOLVE FINISH C*** c*** c* c*** This program displays the results of the finite element solution Save the program Exit PREP7 mode Execute the program .Exit /SOLU mode C*** C*** /POST1 PLDISP PLNSOL,S,X PLNSOL.S.Y FINISH EXIT C*** c*** Display the displacement results Display the stresses in the x-direction Display the stresses in the y-direction Exit /POST1 mode Exit ANSYS 

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.831.1-0087030/manifest

Comment

Related Items