Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A laboratory and clinical study on vitreous fluorophotometry Pang, Kian Tiong 1986-12-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata


UBC_1986_A6_7 P36.pdf [ 7.78MB ]
JSON: 1.0085293.json
JSON-LD: 1.0085293+ld.json
RDF/XML (Pretty): 1.0085293.xml
RDF/JSON: 1.0085293+rdf.json
Turtle: 1.0085293+rdf-turtle.txt
N-Triples: 1.0085293+rdf-ntriples.txt
Original Record: 1.0085293 +original-record.json
Full Text

Full Text

LABORATORY and CL-INIGALSTUDY CD X I  VITREOUS  F*I_.UOROF»HOTOMETRY by j^PANG, KIAH TIONG  B.Sc., The University of Toronto, 1983 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF , MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Physics  We accept t h i s t h e s i s as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA A p r i l 1986 €> Pang, Kian Tiong, 1986  In presenting  this  thesis i n partial  f u l f i l m e n t of the  r e q u i r e m e n t s f o r an a d v a n c e d d e g r e e a t t h e  University  o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make it  freely  a v a i l a b l e f o r r e f e r e n c e and s t u d y .  agree that permission f o r extensive for  s c h o l a r l y p u r p o s e s may  for  financial  of  P H Y S I C S  The U n i v e r s i t y o f B r i t i s h 1956 Main M a l l V a n c o u v e r , Canada V6T 1Y3 Date  DE-6  (3/81)  3 0 t  h  April  1986  Columbia  my  It is thesis  s h a l l n o t be a l l o w e d w i t h o u t my  permission.  Department  thesis  be g r a n t e d by t h e h e a d o f  copying or p u b l i c a t i o n of t h i s  gain  further  copying of t h i s  d e p a r t m e n t o r by h i s o r h e r r e p r e s e n t a t i v e s . understood that  I  written  A B S T R A C T The o p t i c a l ,  e l e c t r o n i c sensoring and data acquisition systems  were assembled and the software developed f o r a vitreous fluorophotometer which was then c a l i b r a t e d and used to quantify the i n t e g r i t y of the  blood-retinal  b a r r i e r i n a p i l o t study of d i a b e t i c  retinopathy  and multiple s c l e r o s i s compared to normal controls. Breakdown of the blood-retinal b a r r i e r was quantified by measur i n g fluorescence i n the vitreous at standard time i n t e r v a l s over one hour  following  plasma  dye  Leakage  concentration  was expressed  t r a t i o n at 3mm The of  intravenous i n j e c t i o n was measured  of  sodium  fluorescein.  throughout  the  as a penetration r a t i o of the average  from the r e t i n a to the t o t a l plasma dye concentration.  by  showing values i n approximate agreement with  pathy severity, thus confirming the findings of previous Of 16 multiple s c l e r o s i s subjects, difference  activity. active  between  Abnormally  periphlebitis.  a c t i v i t y categories or  high A new  penetration  level  r a t i o was  of  retino-  signifi-  of  current  associated  finding was the presence of  the  observers.  r e s u l t s showed no  high leakage i n 2 subjects showing no ocular signs of jects  concen-  r e s u l t s from d i a b e t i c subjects showing well defined stages  instrument  with  abnormally  disease.  Sub-  without or with i n a c t i v e p e r i p h l e b i t i s showed breakdown of the  blood-retinal showing the  procedure.  retinopathy severity demonstrated the proper functioning  cant  The  barrier  comparable i n severity to  no or mild retinopathy.  diabetic  subjects  The vitreous d i f f u s i o n constant  dye f o r normal controls and multiple s c l e r o s i s subjects was  s i g n i f i c a n t l y d i f f e r e n t from that i n water.  of not  T A B L E  OF*  C O N T E N T S  ABSTRACT LIST OF TABLES LIST OF FIGURES ACKNOWLEDGEMENT  i i v vii ix  Chapter I INTRODUCTION 1.1 Vitreous Fluorophotometry 1.2 Blood-Retinal Barrier 1.3 Applications 1.4 Multiple S c l e r o s i s 1.5 Aim Chapter II THEORY AND 2.1 2.2 2.3 2.4 2.5  ALGORITHMS Systematic Errors Theory The C-V Group The L-A Group Other Methods  1 1 3 4 4 6 7 8 12 18 21 24  Chapter III THE APPARATUS 3.1 Sodium Fluorescein 3.2 The Hardware 3.3 Other Material 3.4 The Software  27 27 30 39 42  Chapter IV CALIBRATIONS AND PROTOCOL 4.1 Instrument Preparation 4.2 Subject Preparation 4.3 Blood-Plasma Preparation  57 57 63 67  Chapter V ANALYSIS AND DISCUSSION 5.1 C l a s s i f i c a t i o n s 5.2 F-Numbers 5.3 Intra-ocular Lengths 5.4 Plasma C u r v e - f i t s 5.5 LLoD 5.6 Autofluorescence 5.7 P r o f i l e s 5.8 D i f f u s i o n Constant 5.9 Penetration Ratio  68 68 70 72 75 77 78 82 89 95 iii  5.10 LUND. BAS Results 5.11 Other Parameters  104 106  Chapter VI CONCLUSION  109  APPENDIX A COMPUTER PROGRAMMES A.l DAS.PRN A. 2 SCANMENU. BAS A. 3 VITSCAN.BAS A. 4 PLASCAN. BAS A. 5 BATCHRUN. BAS A. 6 REDUCE. BAS A. 7 B/G.BAS A. 8 MINUS. BAS A. 9 SUDATA.BAS A. 10 BLOOD. BAS A. 11 C/VAZ.BAS A. 12 SLOPES. BAS A. 13 PLOT. BAS A. 14 DRAW. BAS A. 15 LUND. BAS A. 16 F i l e Formats A. 17 Sample CONSENT FORM  I l l I l l 113 114 116 117 118 119 120 120 123 127 130 133 135 137 140 142  APPENDIX B MATERIAL USED B. 1 E l e c t r o n i c s B. 2 Equipment B. 3 Model Eye  143 143 144 145  APPENDIX C CALIBRATION RESULTS C. 1 Pod-DAS C.2 Logarithmic Amplifier C. 3 pH Dependence C.4 R/M-Log Amp-DAS C.5 Attenuation C.6 Performance Data C.7 Model Eye Scan P r o f i l e  146 146 148 149 151 153 156 157  APPENDIX D GLOSSARY  158  APPENDIX E ABBREVIATIONS USED  160  REFERENCES  162 iv  L - I S T  1.  OF*  T A B L E S  Average F-numbers using a Gullstrands emmetropic model eye  17  2.  Distribution of subjects  68  3.  Detail subject classifications  69  4.  Average F-numbers and their S.D  71  5.  Average lengths of the intra-ocular media  72  6.  Significance levels for intra-ocular lengths tests  74  7.  Results of linear, least-squares f i t of axial length, and lens thickness in mm to the subjects's age  74  8.  Plasma f i t s  75  9.  Effect of the f i r s t blood sampling time, t * 1 '  77  10.  Average LLoDs and S. D. s  77  11.  Lens autofluorescence,  80  12.  Autof luorescence, S. D. vs DRP states  80  13.  Autof luorescence, S. D. vs MS states  81  14.  D averaged over a l l measurement scans  90  15.  Tests of D vs age and duration  91  16.  Diffusion constant, D by sex-disease states  91  17.  PR3 averaged over a l l 55-70 minute scans,  S.D. vs age  after background subtraction only 18.  PR3 Average • / - S.D. of the various groups  19.  Significance level to reject H* between diabetic and normal groups v  96 98 100  20.  Significance level to reject H1 betveen MS and normal groups  21.  Significance level to reject H* between diabetic and MS states  101 102  22.  Results of LUND. BAS  105  23.  Electronic components of blue circuit board of the DAS.. 143  24.  Results of Pod calibrations  147  25.  Results of Log Amp test  148  26.  Water and buffer sample differences  149  27.  Calibration results of the R/M-Log Amp-DAS  152  28.  Concentration and gradients of attenuated samples  155  29.  Performance characteristics of the VF system  156  vi  L I S T  1.  OF"  F I G U R E S  The "diamond" i s the intersection of the beam and the probe  2  2.  Diamond dimensions and effects  3.  Diamond displacement vith s l i t lamp translation  16  4.  Variation of F vith s l i t lamp movement  17  5.  Different scanning axes  19  6.  Structural formula of Sodium Fluorescein  27  7.  Excitation and de-excitation peaks of fluorescein in blood  8  28  8.  BLOCK Diagram of the VITREOUS FLUOROPHOTOMETER  31  9.  The S l i t Lamp  32  10.  Excitor f i l t e r holder-slide and intensity monitor  34  11.  Pod asembly and circuit  35  12.  Pin diagram of the blue circuit board in the DAS  36  13.  F i l t e r transmission profiles  40  14.  F i l t e r Overlap  41  15.  FLOWCHART for REDUCE. BAS  48  16.  Overhead viev of the angles at which the s l i t lamp was set for LEFT eye scans  62  17.  Localizing techniques  64  18.  Flowchart of the scanning procedure  66  19.  Plot of F-number results  71  20.  Intra-ocular lengths  73 vii  21.  Autof luorescence  79  22.  Bolus effects  83  23.  Sample profiles  84  24.  Comparison of MS profiles  87  25.  Comparison of diabetic subjects' profiles  88  26.  Overall comparison  89  27.  Diffusion constant vs age  93  28.  Diffusion constant vs duration  94  29.  Penetration Ratio  97  30.  Cross-section of the model eye  145  31.  Calibration curve of the Pod-DAS  146  32.  Performance of the Log Amp  149  33.  pH dependence  150  34.  Calibration curve for R/M-Log Amp-DAS  151  35.  Refraction at sample c e l l surfaces  153  36.  Attenuation in sample solutions  154  37.  Model eye sample profile  157  viii  A C K N O W L E D G E M E N T  L A M E N T "Pangs of hunger, Pang?..." No! Pangs of loneliness. Pangs of being d i f f e r e n t ; Pangs Of behaving d i f f e r e n t l y . Pangs of being misunderstood. Pangs of having d i f f e r e n t ideas; Pangs of having d i f f e r e n t ideals. Pangs of being misunderstood. Pangs Pangs Pangs Pangs  of of of of  being i n a d i f f e r e n t vorld. wanting to be the same; yet. needing to be understood to be d i f f e r e n t . being misunderstood.  Pangs Pangs Pangs Pangs  of of of of  alvays d i f f e r e n t interpretations. very d i f f e r e n t experiences; a d i f f e r e n t upbringing; being misunderstood.  Pangs of a small vorld; but, We are of t h i s one world. It hurts.  My Father's hopes are my r e a l i t y .  Thank you, my FRIENDS.  Art thou f o r something rare and profitable? Wouldst thou see a truth within a fable? From "A Pilgrim's Progress" by John Bunyan.  I  -  I N T R O D U C T I O N  1.1  Vitreous  Fluorophotometry  Vitreous Fluorophotometry <VF) i s a c l i n i c a l research technique f i r s t described by Maurice in 1963 [11. a  non-invasive,  standardized,  Its objective i s to provide  reproducible procedure for examining  the integrity of the blood-retinal barrier. I n i t i a l uses included the investigation disease  of  which  diabetic retinopathy  can cause blindness.  (DRP),  a  retinal  vascular  Several other retinal  vascular  diseases have subsequently been investigated. VF i s a method of sampling the vitreous close to the retina assess  the state of intactness of the tissue.  fluorescent into  a  In the procedure,  dye called sodium fluorescein i s injected  to a  intravenously  subject and i t s entry into the vitreous i s measured by pro-  jecting a beam of light into the vitreous compartment and monitoring the  resulting fluorescence.  A profile of the amount of fluorescence  is established by scanning Blong the axis of the eye. By relating the intensity of fluorescence to the amount of dye present at each  posi-  tion along a scan, the total mass of fluorescein that has entered the vitreous  i s then a measure of the permeability of the blood-retinal  barrier. The basic components of the fluorophotometer are: a) A source of focussed light to excite the dye. b) A probe to detect the amount of fluorescence. c) A photomultiplier/radiometer system to convert the signals. d) A data acquisition system to store the converted signals. 1  The  l i g h t source usually employed i s a tungsten bulb that  can  be varied i n i n t e n s i t y . I t s output into the eye i s directed through a focussing system of lenses, and a s l i t . probe consists of a f i b r e optic conduit that i B placed  The the  of the slit-lamp microscope objective 121.  focus  tioned  at  section  I t i s posi-  an angle to the beam so that i t i s focussed on  of  the beam,  detecting the fluorescence from  at  a  the  cross"inter-  secting" volume called the "diamond". (See Figure 1.) The metric system.  f i b r e optic probe conducts the c o l l e c t e d l i g h t to a radio-  detection Recording  system which then outputs to  a  data  acquisition  and/or data-storing device(s) then store the con-  verted signals.  Figure 1.  The "diamond" i s the i n t e r s e c t i o n of the beam and the probe.  2  1.2  B l o o d - R e t i n a l In  the human eye,  molecules  B a r r i e r  there are two barriers to the  transfer  into and out of the vitreous and aqueous media.  These are  the blood-aqueous barrier (BAB) and the blood-retinal barrier The  BRB is a situation of restricted permeability between the  and  the  retina.  epithelium  (RPE)  It functions at the level of the and the retinal vessels.  retinal  Restricted  of  (BRB). blood pigment  permeability  serves to maintain the regulated physical and chemical environment of the retinal neural tissues, i . e . , homeostasis of the retina. The RPE, nuous,  considered the outer BRB C3], forms a uniform, conti-  single layer of cells united laterally by zonulae  occludens.  (See Appendix 0.) The retinal vessels forming the inner BRB 131, lined  by  a  continuous layer of non-fenestrated  endothelial  are cells  which are joined near their luminal surfaces by zonulae occludens, junctional  complex  in which there i s complete fusion of  leaflet of neighbouring c e l l membranes. of  the  outer  The integrity or "tightness"  the BRB may be compromised by disease processes which affect  of i t s components - the RPE,  a  the retinal capillaries,  arteries  any and  veins. In niques  the  present study,  VF instrumentation and analysis  tech-  were developed and applied to an investigation of the BRB in  multiple sclerosis (MS),  an application that had not previously been  investigated. The results were compared to normal controls and graded severities of DRP.  3  1 - 3  App  1 i c a t i.oris  In VF studies of diabetes m e l l i t u s , leakage  of  suggested down  researchers found abnormal  dye even when there was no v i s i b l e  DRP [4,53.  This  that VF would be useful t o detect the onset of the  break-  of the BRB.  D i a b e t i c persons with w e l l established DRP showed  large amounts of the dye i n the v i t r e o u s .  These r e s u l t s implied that  VF could p o s s i b l y be used t o monitor the leakage component of s u b c l i n i c a l retinopathy progression t o severe stages. The  technique has since been used t o study other r e t i n a l  c u l a r diseases, f o r example, the  vas-  hypertension and pars p l a n i t i s [ 6 ] , and  e f f e c t s of drugs such as s u l i n d a c which has r e c e n t l y been  found  to be e f f e c t i v e i n reversing abnormal e a r l y leakage i n diabetes [71.  1.4  Multiple  Sclerosis  The l i m i t e d understanding of the disease processes i n MS a r i s e s from  the lack  progression.  of good c o r r e l a t i o n s between  relapses  and disease  This imposes d i f f i c u l t y i n the development of an e f f e c -  t i v e treatment. The described appears  perivenular c e l l i n f i l t r a t e i n the cerebrum that has  been  as  [81,  an e a r l y event i n the formation of an MS plaque  similar  t o the r e t i n a l perivenular i n f i l t r a t e found i n the  eye. In laboratory studies of t i s s u e preparations it  (immunoperoxidase),  was found that there was abnormal r e t i n a l venous permeability i n  areas effects  with  and without v i s i b l e p e r i p h l e b i t i s  [91,  and that  such  were much more frequent and extensive than previously c o n s i -  dered. E n g e l l and Andersen [101 estimated that almost a l l MS p a t i e n t s 4  would develop retinal periphlebitis at some point in their If  this is correct,  periphlebitis, larger  lifetime.  i t can be expected that following the onset  there  is  breakdown of the BRB (to  molecules  than fluorescein) which persists despite c l i n i c a l and  of  much histo-  pathologic resolution of the lesions. Clinically, retinal periphlebitis in MS may affect one, several or  all  of the retinal veins and appears either as an active  lesion  with patches of fluffy white haziness surrounding the veins or as  an  inactive venous sclerosis when there i s halo sheathing. The course is mild,  asymptomatic and transient with activity lasting weeks, months  or up to two years. Resolution leaves no sequelae or else replacement sclerosis. Inflammatory activity can be confirmed c l i n i c a l l y by fluorescein angiography photography which shows leakage, whereas inactive venous sclerosis shows no leakage [11,12]. however, show  a  Photographic information,  is limited by the sensitivity of the emulsion and would not subtle breakdown of the BRB that might persist  after  reso-  lution of inflammation. A significant proportion of MS patients with active bitis  also  show  abnormal brain scans when compared to  inactive  periphlebitis [13].  activity  at  periphlethose with  An explicit relationship between  the two sites - the BRB and the blood-brain  barrier  with regards to relapses and disease progression has not been tigated.  5  the  inves-  1 . 5  A i m  The o b j e c t i v e s of t h i s study are the f o l l o w i n g : (1) changes regard  To  use  VF as a s e n s i t i v e system to study  i n MS cases showing a c t i v e and i n a c t i v e to q u a n t i t a t i v e differences i n leakage.  possible  BRB  periphlebitis  with  A p a r a l l e l study  on  persons with diabetes i s conducted as a c o n t r o l on the performance of the VF system. (2) relation  To  document  the frequency and s e v e r i t y  of  to the c l i n i c a l grading of the c e r t a i n t y of the  leakage  diagnosis,  and i n r e l a t i o n to the the standard c l i n i c a l a c t i v i t y categories the  in  with  aim of e s t a b l i s h i n g r e l a t i o n s h i p s between the ocular and c e n t r a l  nervous system a c t i v i t i e s , p a r t i c u l a r l y relapses and disease progression. (3) technique  To assess the u t i l i t y of the procedure as a in  the diagnosis of MS,  and as  an  i n d i r e c t monitoring  method of grading the c e n t r a l nervous system a c t i v i t y .  6  non-invasive  I I -  In rated.  T H E O R Y  A N D  this  the details of the VF technique are  chapter,  A L G O R I T H M S  elabo-  The algorithms proposed by two groups of investigators  were  closely followed in this experiment in order to compare results.  The  methods of these two groups are detailed in the following sections. The f i r s t section explains the "systematic errors" inherent each  VF scan.  ring  instruments,  in  These errors arise from the limitations of the measuas well as from the complex  system of the human eye.  Thereafter,  optical,  biological  there is an explanation of the  earlier, more basic models used in analysis. The methods  second  of analysis that were used in this study.  accommodate and  set of sections discusses the two  more  elaborate  The algorithm  the "systematic errors" was proposed by  J.G.  co-workers and is referred to as "The C-V Group".••  to  Cunha-Vaz The state  of the BRB i s then embodied in a single number called the Penetration Ratio.  "The L-A Group" described a more  solution  mathematically  to find the Permeability and Diffusion  involved  coefficients.  This  algorithm is due to H. Lund-Andersen and co-workers. In the development of the VF system, col  and  alterations to the proto-  the algorithm necessarily occur because of  procedures and instrumentation.  differences  in  Some modifications to the methods of  the two groups are discussed in Section 2.5.  ** " . . . Group" refers to a general geographical distribution of the various investigators, and i s also used to distinguish between those using one method of analysis and those using the other. 7  2-  1  Systematic  Errors  Figure 1 shows that the "diamond" i s of f i n i t e it  i s moved across an interface from a compartment of high dye  centration obtained  into  effect  a  compartment without dye,  from the empty compartment.  choriod-retina  <CR)  a non-zero  signal  One such interface i s at  and the vitreous,  where t h i s  This  As con-  non-zero  is the  signal  i s especially s i g n i f i c a n t during the f i r s t few minutes  the introduction  after  of fluorescein into the blood.  "false"  signal  i s due to the depth or  diamond as i l l u s t r a t e d below. dye-free compartment. or  dimensions.  length  of  the  It persists f o r some distance into the  This effect i s sometimes c a l l e d the " t a i l i n g s "  "spread function" due to the associated  peak because i t s strength  depends on the peak signal at the interface [14.].  Choriod I Retina  Posterior Vitreous  0  V :  8.2512*11  D ' d*cos 8  g  + ?*cos 9  t  Peak  a. Estimates of Axial Resolution and diamond volume. Figure 2.  b. The " t a i l i n g s " due to the choroid-retinal peak [18].  Diamond dimensions and e f f e c t s .  8  There  are other sources of "false"  close to the retina. the  retina,  signals,  especially  One effect is halation [15].  very  When focussed on  the edges of the s l i t are not distinct.  This i s due to  the transparent depth of the retina (about 0.5mm), and to scattering. There i s also a possible dependence on retinal pigmentation [161. There may be signals from light reflected off the walls of vitreous  cavity although i t is unlikely that,  away from the retina,  the reflected light intensity can be sufficiently high, direction  of  the probe.  the  This is also apparent  from  or be in the the  relative  volumes of the diamond and the vitreal cavity. As consider  tailings are strongest near the retina, data  that are collected at a more remote point  spread function is small. peak  the solution is to  However,  where  the  the retina and i t s associated CR  must always be included in any scan as they constitute a  refe-  rence zero-position by which displacement can be measured. The  distance  from  the retina within  which  accepted depends on the dimensions of the diamond.  data  cannot  be  The in vivo axial  resolution (AR) is defined as the distance from the CR peak where the signal must  i s a small not  detector.  fraction of the peak [141.  The fraction  be so small that the signal is at the noise level  chosen of  the  A practical definition of AR i s the ratio of the signal at  a fixed distance from the retina to the CR peak signal [171. Figure 2a estimates the AR and the volume of the diamond.  They  depend on the angle between the directions of the excitation beam and probe,  the width of the s l i t ,  and the diameter of the probe.  parameters must be varied until an optimized AR is attained 9  These  [181.  Reducing  the s l i t width and/or using a smaller probe  need not necessarily result in better ARs,  diameter  as a reduction of  either  or both reduces the amount of fluorescence detected. The noise of the detection system then limits the reduction that is possible. However, it  is necessary to have signals distinctly above the noise levels so  that the data can be analyzed with greater confidence. slit  width  cannot be measured ( i . e . ,  calibrated)  Also, a small  with  precision.  (Refer to Sections 3.2, 4.1 and Figure 17.) Increasing  the  angle shown in Figure 2a improves the  reduces the volume of the diamond. the  diameter  to  AR and  However, this angle is limited by  which the pupil of the eye  can  be  dilated.  cornea's curvature and i t s varying thickness also distort the beam  and s l i t ,  formulae  in  finite  thereby reducing the probe's focus on the beam.  Figure 2a are therefore approximations as the  The  The  in  vivo  among  sub-  dimensions of the diamond cannot be measured directly. The jects.  maximum dilation diameter of the pupil varies  When  diameter,  the  the  pupil  procedure  cornea  curvature,  placed  on  the  cannot be dilated to a cannot be used.  however,  cornea.  minimum  The power of  can be offset by a  acceptable the  convex  plano-concave  The plane surface of the lens  lens  provides  a  "window" to view the fundus. (See Figure 3.) Figure 3 also shows the importance of AR. The angle between the beam  and  changes  the probe in the vitreous varies during  with i t .  scan.  The AR  (See Figure 2a.) The AR i s larger in the posterior  vitreous than in the anterior segments. causes  a  more severe tailings,  As the larger AR near the CR  i t s optimization i s thus v i t a l to 10  the  instrument's performance. Many  biological  fluorescent. crystalline  They lens  fluorescent. tailings),  fluoresce  at certain incident  to  be  auto-  wavelengths.  and the cornea (as well as the retina)  They BS  substances (tissues) are known  give off false signals in their  The  are  vicinity  auto(i.e.,  well as absorb part of the excitation beam which must  pass through them.  when  The  autofluorescence  of the crystalline lens  the  vitreous i s to be scanned.  is  unavoidable  It depends on the age  of  the  subject as well as the disease process [19]. Cataracts (or opacities) in  the  lens also cause loss of input intensity,  which  limits  the  application of VF to eyes with clear media. In  the later scans,  there is usually a high concentration  dye in the anterior chamber which can cause a loss of incident due  to absorption,  expected  levels  injection  are  or stray signals from scattering.  of fluorescein in the aqueous at 60 sufficiently  small that they do  not  of  light  However, minutes  the after  attenuate  the  excitation beam appreciably. (See Appendix C.5.) •ther sources of error are related to the apparatus. In Section 3.1, of  where the properties of fluorescein filter  overlap will be mentioned.  are discussed,  Problems in  the  the problem radiometric  system such as dark current noise due to random photon events in photomultiplier  tube  (PMT) are minimized by  constantly  the  checking  instrument zero adjustments. All sity  the above sources of variation are affected by the  of the excitation beam,  inten-  which affects the amount of scattering 11  and therefore autofluorescence, zation  of  tailings,  and AR.  the apparatus involves the adjustment of  Hence, all  optimiparameters  which contribute to the quality of the data acquisition.  2.2  Theory The permeability of the BRB i s related to the diffusion of  dye across i t .  the  Concentration differences and electric potentials are  some of the forces driving the diffusion phenomenon. In the f i r s t two hours  after the injection of the dye,  passive diffusion governs i t s  penetration through the BRB from the blood into the vitreous [ 2 0 , 2 1 ] . This means that Fick's Law, driving force, the BRB, known.  in which a concentration gradient i s the  can be applied. Hence, to measure the permeability of  the change in the blood-dye concentration over time must be The concentrations of dye on both sides of BRB may  related  then  be  by a proportionality constant which represents the permeabi-  lity. The simplest mathematical model is that of a plane retina [ 2 2 ] . The equivalent one-dimensional problem i s (1)  D * d f t c(r,t)/dr B  = dc<r,t)/dt ,  where the concentration, c, i s position-dependent and time-dependent, r is the distance from the middle of the vitreous. D i s the diffusion coefficient which is assumed to tbe independent of c,  and hence, may  alternatively be called the diffusion constant. The  assumptions of such a model are that (a)  fluorescein  can  only diffuse towards the middle of the vitreous after penetrating the 12  BRB plane,  (i.e.,  a one-directional transport process); and,  (b) the  boundary conditions, which are:  (2)  dc(0,t)/dr = 0 ,  (3)  D * dc(a,t)/dr = P c < t ) , l  p  where the retina i s at a distance a from the mid-vitreous, the concentration of fluorescein i n the blood at time t. d i f f u s i o n constant, P , 1  and c" i s Like D, the  the "permeability c o e f f i c i e n t " i s independent  of c, and i s referred to as the permeability index [23]. The solution to Eq. 1, using Eqs. 2 and 3 i s  c(r,t) dr (4)  r -  t  T • t  c T  P  1  relates  trate  (T)  dT  • a  the t o t a l amount of dye i n the blood available to  the BRB at the post-injection (p.i.) time,  amount area  p  i n the vitreous at t.  Mathematically,  t,  to  the  and  the  retina.  total  the numerator i s the  under the concentration p r o f i l e (taken at t) between  vitreous  pene-  The denominator i s the  area  the  mid-  under  the  plasma-fluorescein concentration p r o f i l e up to t. Several problems arise i n solving Eq. at  the  BRB  transport [20,21], fact,  are not simple.  phenomena"  It has been  4. The processes at work established  that  drive solutes against concentration  "active gradients  The forces driving these active transport processes are, i n  greater  than those for passive d i f f u s i o n by about  13  31  times.  They  come  outward  into effect after the f i r s t two hours p . i .  Unless  active processes are to be studied the last scan is  these usually  taken at approximately one hour p . i . The time taken for the dye to reach the eye depends on the site of injection, is  injected  peripheral  e.g.,  the dye appears at the retina 10s sooner when i t  into a carotid artery than when i t is injected vein  quickly  also  profile  of  in the arm [24,25].  affects  Injecting the dye  the time of appearance at  the  the bolus which is attenuated even after a  into  slowly  BRB and fast  a or the  intra-  venous injection due to mixing with blood. Venous  blood  samples  obtain the plasma profile. several the  minutes p . i .  method  within rule. t'11,  the  are drawn throughout the The f i r s t sample  is usually taken  e.g.  more samples are  60 minutes i f the area is found  using  the  to  after  and the number of samples required depends  of solving the integral,  Assumptions  procedure  on  needed  trapezoidal  must also be made about the profile between 0 and  the f i r s t sampling time when the latter method is used [261. Curve-fitting  techniques  may be used in  solving  the  plasma  integral. In pharmacokinetic studies [27], i t has been found that the time-course  of  two-compartment  fluorescein in the blood is best approximated model  of  mixing.  This requires a mode of  fitting a sum of two negative exponentials to the data [28,29] includes  by a curvewhich  a fast and a slow decay in the levels of fluorescein in the  blood. (Refer to Section 3.1.) The limit,  integral in the numerator in Eq. 4  assumes  in i t s  lower  that dye penetrating the BAB and dye penetrating the BRB have 14  not  mixed within one hour p . i .  127]  Tailings of the CR peak  t r i c t the upper limit of the integral as previously  mentioned.  resHow-  ever, i t can be expected that most of the fluorescein in the vitreous i s in the vicinity of the retina as will be explained. Attenuation anterior  of  chamber  was  the  excitation source by fluorescein  briefly mentioned in  Because  of  i t s small volume,  chamber  by  way  distribution  the  the dye quickly  of the i r i s and c i l i a r y body.  (because of i t s small volume),  previous fills  the  section.  the  Assuming  in  anterior a  uniform  the attenuation may be  expressed in the form of the Beer-Lambert law: (5)  c(measured)  =  c(true) * exp< - b » cft * d ) .  This assumes that only the concentration, of depth, segment the  c", in the aqueous chamber  d, attenuates the excitation beam. The dye in the anterior of the vitreous (adjacent next to the posterior  lens)  and  lens autofluorescence  surface  are not included.  attenuation (or extinction) coefficient [183.  b  is  This attenuation  less than 10% for c"<1000' at the one hour scan.  of the is  (See Appendix  C. 5. ) Figure in  the  axis,  3  eye.  below shows how rays are refracted at the interfaces  As the diamond i s moved anteriorly along  the  optical  the angles of incidence change at the various surfaces so that  AR is also position-dependent. Figure 3 lamp  also demontrates that a 1-mm translation of the  slit  does not correspondingly produce a 1-mm diamond displacement in  the medium in which i t is focussed. 15  It is thus necessary to translate  Figure 3.  slit  lamp  Diamond displacement v i t h s l i t lamp t r a n s l a t i o n .  translations to displacements of the  l a t t e r are not d i r e c t l y  diamond  since  the  measurable.  S l i t lamp translations, d are related to diamond displacements, x, by  [30] x  The  "F-number*  face. for  represents the effects of refraction at each  F i t s e l f depends on x.  a model eye by Krogsaa,  the  compartments  the  diamond  = F * d .  Table 1 and Figure 4 show the et a l [30].  interresults  Note that F i s d i f f e r e n t i n  because there are'fewer interfaces to traverse  i s moved towards the cornea.  Also,  demonstrating the power of the ocular system.  16  F>1 i n  as  a l l cases  Model Eye Distances in mm  Compartment  Aqueous Chamber Crystalline Lens Vitreous Chamber  Table 1.  3.60 3.60 16.97  S l i t Lamp Movements in mm  F-number  2.49 2.20 11.75  1.45 1.64 1.44  Average F-numbers using a Gullstrands emmetropic model eye. [301  1.6 -  1.5 -  1.4 •  AQUEOUS  LENS  VITREOUS  SLIT LAMP TRANSLATIONS Figure 4.  Variation of F vith s l i t lamp movement. [301 17  2 .  3  T h e C —V  A simple  G r o u p  way to quantify the permeability of the  BRB is  to  measure  the amount of fluorescein in the posterior vitreous segment.  However,  different instruments with different sensitivities and ARs,  measure different strengths of CR tailings.  Therefore, a number that  quantifies the permeability must be independent of instrument  diffe-  rences. To injection  offset errors due to CR peak tailings, scan  i s subtracted from a l l other  a background,  post-injection  pre(p.i.)  scans. Lens autofluorescence i s also eliminated by this subtraction. The subtraction i s carried out by aligning "landmarks" such the lens and CR peaks of each scan.  This method,  however,  does not  consider possible shifts in the CR peak position with time, i . e . , CR peak  may  not be the true position of the retina  in  as  the  the later  scans. As fluorescein is continuously removed from the blood,  a point  may be reached when the signal from the posterior vitreous is greater than that from the CR (in high leakage cases). appears  in front of the retina.  Also,  That is,  the CR peak  due to different fixation or  scanning axes, the distances between peaks may vary. An The  alternative method is to align only the  retinal  retina may be located visually at the start of each  assumes  scan.  This  that the starting points of each scan are at the same  loca-  tion on the retina. starting landmark. different  position.  The macula,  for example,  may be used as such a  An advantage of this method is that errors due to  alignments  (shown in Figure 5b) are smaller near the CR.  18  a. Different starting points.  Figure 5.  However,  b. Macula alignment.  Different scanning axes.  the data collected from the anterior half of the eye may  no  longer be analyzed vith confidence. The d i f f i c u l t y i n this method i s the accuracy with which the CR can  be located due to halation,  scattering,  and the centring of  a  small s l i t on a larger probe. In order to account for the possible e f f e c t s of CR t a i l i n g s the  l a t e r scans,  The  dye  within  a "bolus" scan i s made within 3 to 5 minutes p . i .  i s not expected to have penetrated the BRB the f i r s t 10 minutes p . i .  strength  of  in  (in most  The bolus scan then  cases)  provides  the CR t a i l i n g s at s p e c i f i e d distances from the  the  retina  after the background has been subtracted. The distance from the retina at which calculations are done usually  the 3-mm  point.  is  It i s expected that the ARs of most i n s t r u -  ments are smaller than t h i s [311. 19  Thus,  CR t a i l i n g s should not  be  very significant at this point.  Any point where the tailings are not  substantial can be chosen. However, to study leakage from the BRB, i t is  important to scan close to the retina.  Therefore,  3mm from  the  latter is a suitable choice (after AR considerations). The  CR tailings are subtracted from later measurement  scans**  only i f the former are significant. Ishimoto, et al [323, suggested a criterion  for  this  bolus CR correction with a  recommendation  for  implementation i f (6) The  (CR Peak value)/(3-mm Vitreous value) > 10 . above  equation was based on the commercial  fluorophotometer.  Fluorotron*  The condition may be different for other  Master instru-  ments. The before  the  measurement  scans are taken at specified  outward transport processes  alignment and background subtraction, necessary),  become  time  intervals  significant.  After  (and bolus CR correction,  if  the average value of the dye concentration around the 3-  mm position is found [323. gral up to the p . i .  The result is divided by the plasma inte-  time of the scan. The final result is called the  penetration ratio, PR3, of units,  s" 1 .  The advantage of using PR3 instead of P1 in Eq. 4,  i s that the  3-mm value from the retina may be used for comparison between different patients and instruments.  There are no problems of CR tailings,  and i t is not required to integrate very close to the CR. ** A l l scans other than the pre-injection and bolus scans are referred to as measurement scans. 20  The points  averaging  E323.  i s carried out between the 2-mm  and  the  4-mm  This reduces errors usually caused by random f l u c t u a -  tions i n the data which can persist,  even a f t e r subtraction,  penetration r a t i o at the 3-mm  PR3*.  point,  for  a  The averaging also dimi-  nishes the errors due to the alignment problems mentioned before.  2 .  4  T h e Much  this  I_ — A  of  group  G r o u p  the scanning and data-correction techniques  are  the  same as that of the C-V  similar instrumentation). elaborate, eye  Group  (by  used  by  virtue  of  However, instead of using PR3 or P ,  a more  1  mathematical model of the transport of fluorescein i n the  i s constructed to estimate the vitreous d i f f u s i o n  and the permeability constant,  P, of the BRB  constant,  D,  E33],  The analysis involves solving the d i f f u s i o n equation,  (7  )  as  V • (D7c) a  = 6c/6t ,  boundary-value problem d i r e c t l y .  approach  is  circulation.  The d i f f i c u l t y continuously  in  such  removed  an  that  fluorescein i s being  from  This  means that the transport problem i s a "transient"  one. Several  assumptions of the previously  model of the C-V from retina that  the is the  BRB  group apply to t h i s model.  have not mixed uniformly  mentioned  plane-retina  (1) Dye from the BAB  at 60 minutes  and  p.i.  (2)  The  assumed to be spherical with radius of curvature  a  so  towards the centre of  the  d i f f u s i o n i s r a d i a l l y inward,  vitreous chamber E233.  Furthermore,  21  i f D i s independent of c, then  Eq. 7 reduces to an r-dependence only:  (D/r)-{ 2 6/6r • r 6*/6r« } c(r,t> = dc(r,t)/6t ,  (8)  for 0 £ r £ a Due to symmetry, the boundary condition i s  oc(0,t)/6r = 0 .  (9)  The i n i t i a l condition i s c(r,0) = 0 ,  (10)  Thus  for  0 < r < a  f a r , the model assumes that the concentrations  on  both  sides  of the BRB are related through a proportionality  constant,  which  represents  P  reference outer). sion  to  the  the  permeability of the entire BRB.  location of breakdown within the  BRB  makes  no  (inner  or  Thus, at any time, t, the amount of dye available f o r d i f f u -  towards the mid-vitreous,  that has already penetrated  depends on the amount of dye  the BRB.  (ID  - D 6c(a,t)/6r = P-( c(a,t) - C (t) ) ,  (12)  { 1 + <D/P)-6/or ) c(a,t) = C < t ) .  Using the method of Laplace transforms, the solution i s  c' (r,s) = c''<s)  where 22  c(a,t)  But c(a,t) depends on c"(t)  assumption. The b a r r i e r condition i s then  (13)  P  F' <r,s)  by  Q4)  F'(r,s) =  u(r) sinh v(r,s) , § sinh w(a,s) • J"s cosh w(a,s)  <15)  u(r) = aP/r/D ,  (16)  w(r,s) = W<s/D) ,  (17)  for  0 < r S a ,  for  0 < r < a ,  5 = u(a) - /D/a .  s i s the transform variable; prime implies the transformed functions. Transforming  back  to  r e s u l t s f o r small t C33], r e a l part of s,  t-space, However,  a  slowly-convergent  series  i f Eq. 14 i s expanded f o r large  and then inverted, the approximate solution to Eq. 7  is (18)  c" (t-t) F ( r , T ) d T ,  c(r,t) *  T - •  where (19)  F ( r , T ) = u<r)-( (exp<-M<-1) ) - exp(-M(1)*))//(in ) 8  -  5-exp(5 T)•(exp(2g(T)M(-l))-erfc(g<T)+M<-1)) 2  - exp(2g(T)«(l))•erfc(g(T)*M(l))) } , and (20)  W(j) = M(r,T,j) = <a-jr)/(2V(DT)) ,  (21)  g(T)  (22)  = 5/T  erfc(x) = (2/Vn)  j=-l,l  .  exp(-u*) d p ij II •  •  Eq. 22 i s the complementary error function. Eq. 18  may  then be used as the t h e o r e t i c a l solution i n a non-  l i n e a r c u r v e - f i t t i n g calculation ness of f i t may  to the experimental data. The good-  be tested by k - N  (23)  S  8  =  E <cMr(k),t) - c(r (k), t))» • 6 (k)° , k - >  where  23  (24)  e(k)» = max «fl", c*<r(k),t) ) 4  serves as a 'weighting factor', and.  (25)  fi  c*(r(k),t) retina,  a lowest concentration that can be detected . i s the concentration measured at (a-r(k))  of the scan taken at time t.  The index k  order of the data-points along the axis of a scan.  mm  from the  represents the  Non-linear curve-  f i t t i n g i s then carried out by the Marquardt algorithm [ 3 4 ] . A search through  the parameter space of P and D- i s done and the best  determined  2. 5  is  when S* i s a minimum for one set of P and D values.  OtherThe  fit  Methods  algorithms  used  by  other groups  of  investigators are  usually variations of those discussed above. For example, Eq. 19 may be s i m p l i f i e d f o r  computational purposes [ 2 3 ] .  Most  modifications,  however, arise owing to differences i n instrumentation which necessitate differences i n protocol and algorithm. Different  methods  have  been proposed f o r the bolus  CR  peak  correction mentioned i n Section 2.3. Since many investigators use the Fluorotron* apparatus the bolus  Master,  one  method of correction  suggested  f o r this  was to multiply the bolus p r o f i l e (between 2 and 4 mm from  retina) by the r a t i o of the CR peaks of the measurement and the scans [153.  The modified bolus  from the measurement scan.  p r o f i l e was then  subtracted  This seems reasonable as t a i l i n g s  depend  on the peaks causing them. Bursell,  et a l [353 argued that t h i s algorithm over-corrected  24  the errors due to t a i l i n g s because the CR peak values included rescence  from  the  vitreous because of  Figures  1  smaller  than that between CR peaks.  dimensions  the  finite  diamond.  and 2.) The correcting r a t i o should i n f a c t  of  However,  the diamond i n vivo cannot  fluo-  be  (See  slightly  the variations of the  be  determined,  and  the  appropriate correction factor i s unknown. It  has  been found that r e t i n a l blood flow increased by 40  70X i n the t r a n s i t i o n from l i g h t to darkness [36]. not  l i k e l y to a f f e c t most scans,  plasma  integral,  some  Although t h i s i s  except possibly the bolus and  investigators use more  intense  excitation  such  as xenon f l a s h tubes to a t t a i n s i g n a l  above  dark  current noise so that room l i g h t i n g  the  to  the  sources levels  need  of veil  only  be  dimmed and not completely turned o f f [37], For instruments that scan continuously, i . e . , c o l l e c t data continuously  along  the  scanning  path,  the  signals  positions overlap because of the f i n i t e diamond.  from  adjacent  The methods used i n  "smoothing" the signals (integral and c u r v e - f i t t i n g methods) are then important using  [38].  In other instruments that employ the "spot" method  "chopped" or f l a s h excitation sources,  s p e c i f i c points along the scanning axis, These  tvo methods of data c o l l e c t i o n  data are c o l l e c t e d  e.g. at every 1mm  determined by  at  interval.  characteristics  such as AR also determine whether PR3 or PR3» i s to be employed. To compare the r e s u l t s obtained by various fluorophotometers, a set  of instrument c h a r a c t e r i s t i c s i s used to describe  ment's  capabilities.  This  set of performance  data  meters such as angle between the beam and the probe 25  each  instru-  includes directions,  paraAR,  f i l t e r overlap,  lower l i m i t of detection (LLoD), r e p r o d u c i b i l i t y (R)  and error of measurement < EoM). The  i n vivo  detectable)  concentration  Practically,  plus twice i t s standard  i t means that,  background scan, detected  LLoD may be defined as the lowest  f o r example,  detected  deviation [32].  at the mid-vitreous  the LLoD i s the average value of the  i n that region plus twice the standard  (or  of a  concentration  deviation  of  that  average. The ability  sensitivity  of  as the  of the detection system to d i f f e r e n t i a t e changes i n adjacent  concentration  volumes.  I t i s i n v a r i a b l y dependent on the diamond and  the ambient concentration. R  the detection system i s defined  (See Appendix C.6.)  and EoM also have the same  dependences.  Their  definitions  are: (26) (27)  EoM = { c(measured)/c(true) } - 1 ; R = standard  deviation of repeated measurements.  The problem with these d e f i n i t i o n s i s that the i n vivo "true" concent r a t i o n s cannot be determined.  26  I I I .  3 .  1  T H E  A P P A R A T U S  S o d i u m Sodium  F l u o r e s c e i n  fluorescein  phthalic anhydride i n 1872  was f i r s t synthesized from r e s o r c i n o l [391.  and  It i s a very weak dibasic acid with  a molecular weight of 376.27. Its s o l u b i l i t y i s increased as a sodium salt.  Na*  -0  C00-Na*  RESORCINOLPHTHALEIN SODIUM Figure 6.  -  C..H 0,Na« lt  Structural formula of Sodium Fluorescein E40],  Its s u i t a b i l i t y as an indicator i n ophthalmological research i s due to the fact that the peak excitation wavelength (490nm) i s d i f f e rent from the peak emission wavelength (520nm). excitation time i s short:  approximately  4ns.  In addition, the deHence, with a s u i t a b l e  combination of f i l t e r s to separate the two wavelengths of l i g h t , concentration may  be deduced from the amount of fluorescence.  27  the  1  2  4000  5000  6000  7000  Wavelength (AngstrSm Units)  Figure 7.  Excitation (1) and de-excitation (2,3,4) peaks of fluorescein i n blood [25].  The choice of f i l t e r combinations, by  however,  i s made d i f f i c u l t  the fact that the ranges of excitation and emission  shift  wavelengths  towards the red end of the spectrum when fluorescein  i s mea-  sured i n blood compared to measurements i n water solutions [413.  This  e f f e c t may be caused by multiple scattering, absorption and autofluorescence  of the tissues that are scanned.  The optimum  "cross-over*  point f o r the f i l t e r combination should be about 525nm. Fluorescein  diffuses  readily from the blood into  a l l extra-  c e l l u l a r f l u i d s except across the retina (BRB), and the brain brain  barrier).  As  i t i s a weak acid,  i t does not bind  with (or  stain)  normal v i t a l tissues,  media.  In aqueous solution, about 80*/. of incident l i g h t i s converted  to  fluorescent  and i s highly fluorescent  (blood-  radiation [39,403.  However,  the dye only  approximately 2G7. fluorescence when dissolved i n blood. due  to binding  to proteins  (serum albumin) 28  i n alkaline  and  returns  This loss i s  red blood  cell  membranes. Another effect is quenching by the haemoglobin. The absorption  spectrum of haemoglobin is about identical to that of fluores-  cein.  This  there  is a stronger fluorescence as proportinately less of the  is  can be demonstrated in severely anaemic  quenched by the haemoglobin.  ultra-filtration  [40],  fluorescein is bound. that  diffuses  disrupted.  across  This  it  patients  vhere dose  By means of equilibrium dialysis or  is estimated that between 50 to  It is however, cellular  84% of  the unbound fluorescein  membranes  and the  effect must be taken into account  (17%)  BRB i f  it  is  when analyzing  plasma scans. Fluorescein inability lethal this  has  low  toxicity which i s probably  to bind with v i t a l tissues [393.  due  to  its  In animal experiments,  doses were at 2 to 3 grammes per kilogramme body  weight.  VF study and other investigations on human subjects,  the  In dose  administered is calculated at 14"1 body weight using pharmaceut i c a l ^ prepared ampoules of 25% concentration (2.5"1). Fluorescein  is  well tolerated but there are  occasional  side  effects such as transient nausea or vomiting immediately after injection.  Yellowish tinting of the skin lasts for several  injection and the urine is yellow for about two days. tions are rare. Other the pH,  hours  Allergic reac-  (See Appendix A.17.)  important dependences of i t s fluorescent property are on  concentration and temperature. Only in an alkaline medium is  i t s fluorescent property enhanced [39,40].  It was found that pH 7.4  is the level at which the dye fluoresces most efficiently [42]. is  after  approximately  the pH of the cellular fluids of the 29  body  This which  varies l i t t l e . tions  as  Importantly,  veil  the pH of the c a l i b r a t i o n sample  as the buffer f o r d i l u t i n g plasma  samples  solu-  must  be  result  of  s p e c i a l l y prepared. The  dependence  scattering  the ambient concentration i s a  of the incident beam at the focus of the probe.  concentrations, dye.  on  the  Attenuation  2  The? This  cations  describes the instrumentation  diagram of the VF system  with  principal  specific  the" . 1  H a r d w a r e  section  lamp microscope.  of  of the incident beam causes loss of signal at  that were made to the instruments.  The  high  excitation beam cannot penetrate the volume  detector. In t h i s study, the upper l i m i t i s about 0.01  3.  At  and  the  modifi-  Figure 8 shows the block  assembled f o r t h i s study. (Appendix  component of the fluorophotometer i s  B.2)  the  (Figure 9.) The b u i l t - i n power supply (from i n t e n s i t y settings was replaced by a  mains)  regulated  variations i n beam i n t e n s i t y were  d.c.  supply  because  found  to  occur.  These fluctuations were believed to a r i s e from variations  of  the  random  slit  l i n e voltage when the number of users increased  (i.e.,  unregu-  lated l i n e ) . In order to continuously monitor the intensity, cell  was  through bulb,  placed along the path of the beam before i t the s l i t and prism system.  did not block the beam's path.  This c e l l ,  a photovoltaic was  focussed  placed close to  the  This method of monitoring lamp  intensity, which i s dependent on o p t i c a l alignment, was compared with another method which monitors the i n t e n s i t y of the output of the s l i t 30  r  DVM  R e g u l a t e d Lamp Power S u p p l y  Excitor Filter/Intensity Monitor Slide  Tungsten Bulb RETINA S L I T L A M P A S S E M B L Y  Eye-piece 4 Fibre Optic  i,  to  y i  i_  POD:=  CONTACT LENSES  Barrier F i l t e r Shutter  IF L O G  D A S  M  A M P  R /  M  Ribbon Cable to Computer O S B O R N E Dot-Matrix F>R  I N T E R  M I C R O C O M P U T E R  F i g u r e 8.  BLOCK Diagram of the VITREOUS FLUOROPHOTOMETER  31  26  17.  1.  Fixation  2.  Hruby lens  3.  H r u b y lens guide  4.  25  lamp  Forehead  24  Coaxial knob for slit control  and  rotation rail  rest  18.  Socket for slit l a m p  19.  Knob  for  height  housing  adjustment  m i c r o s c o p e slit l a m p  5.  Chin  6.  Knob for chin rest height  rest  7.  A r m clamping  8.  Grip bar  9.  Cord for fixation  adjustment  Socket for fixation Gear box cover  12.  Zoom  13.  Slit tilting  lamp  22.  Main switch and control for  23.  Pilot lamp  24.  Cross-slide  25.  Base plate  26.  1 4 . •• F i l t e r s l i d e Aperture  Power  cord  secondary voltage  ring  16.  21. cord  lever  Slit rotation  Lever for cross-slide m o t i o n , c o a r s e and fine, of t h e t a b l e  lamp  11.  15.  20.  screws  10.  of  assembly  slide  diaphragm  table  Swivel a r m c o n n e c t i n g and  knob  slit  lamp  table  Figure 9 . The S l i t Lamp. (From NIKON Zoom-Photo S l i t Lamp Microscope Bench Type Instructions Manual) 32  output  assembly  microscope to  cross-  illumination using a photocell. Testing  showed  that the l a t t e r method was more  s e n s i t i v e to  variations i n beam i n t e n s i t y . At high i n t e n s i t i e s ( i . e . , running currents through the filament), the  a.c. mains.  high  f l u c t u a t i o n s were noted when used on  These f l u c t u a t i o n s were reduced (halved) when the  regulated supply was i n s t a l l e d . It should be noted that these i n t e n s i t y f l u c t u a t i o n s appear variations  i n the s i g n a l about an average because  continuously  the system i s  e x c i t i n g and detecting the fluorescence i n  overlapping  volumes (because of the diamond). (Refer to Section 2.5.) provided  that the i n t e n s i t y of the bulb does not vary  from the same average value during each scan, tuations  can be  as  Therefore,  significantly  the fluorescence f l u c -  interpreted as deviations about  an  average  con-  centration at any p o s i t i o n i n the scan. It  was also found that maintaining a constant,  illumination tures)  cause  intensities  was d i f f i c u l t because the high currents  high i n t e n s i t y (and tempera-  the bulb i n t e n s i t y to f a l l continuously. could  only be attained a f t e r a  usually about 30 minutes.  However,  Stable  "warming-up"  excessive,  high  period  long periods at high  i n t e n s i t i e s caused a reduction i n the l i f e t i m e of the bulb. The  configuration to monitor the i n t e n s i t y employs the unused  side of the excitor f i l t e r holder-slide. A small s o l a r c e l l was glued to a microscope cover-glass. washer beam  The assembly was,  that f i t t e d i n t o the s l i d e .  i n turn,  glued to a  This method does not monitor the  during a scan because the chip cuts o f f the beam when i t i s i n  operation. 33  Intensity checks are carried out immediately p r i o r to scanning. The  output of the solar c e l l i s measured on an  intensity  LED  voltmeter.  output i s always adjusted to the value (on the  The  voltmeter)  at which calibrations were carried out. S l i t lamp translations are measured by constructing a potentiometer  with  a 10-turn rotary potentiometer/resistor.  F i t t e d with  a  gear on i t s shaft, the pot i s held by an arm attached to the grip bar of the s l i t lamp.  (See Figures 8 and 9. ) The assembly i s referred to  as the Pod. The  gear rests on a rack which i s mechanically coupled to  body of the s l i t lamp. a scan,  the  When the rack moves with the s l i t lamp during  the Pod produces the analogue voltage signals which are read  by the microcomputer. (See Figure 8. ) The  most important modification to the s l i t lamp concerns  oculars  or eye-pieces.  with  fibre  a  One ocular i s replaced by a special  optic conduit at the f o c a l plane  of  the  the  adaptor  microscope  objective. The f i b r e optic c o l l e c t s the fluorescence from the diamond and conducts the l i g h t to the Photo-Multiplier Tube (PMT).  CHIP  Figure 10.  FILTER  Excitor f i l t e r holder-slide and intensity  34  monitor.  Figure 11.  Pod assembly and c i r c u i t .  An e l e c t r o n i c shutter and the b a r r i e r (green) f i l t e r are placed between  the  output of the f i b r e o p t i c conduit  and  the  PMT.  (See  Figures  8.) The PMT then relays to the radiometer (R/M) whose output  voltage v a r i e s l i n e a r l y with the amount of input fluorescent l i g h t . The  R/M has several exponent  adjusting  exponent  option.  between 0 and +100 mV, points  settings,  The AUTO s e t t i n g ,  including  an  which keeps  was found to be unstable at the  AUTOoutputs  "cross-over"  where output voltages greater than +100 mV were scaled  down.  Hence, the R/M i s set i n the (most s e n s i t i v e ) 0-exponent range. To scanned, used. sition small  offset a  high  R/M outputs  logarithmic  when high  concentrations  a m p l i f i e r ( c a l l e d LOG AMP i n Figure  8)  are is  The Log Amp was c a l i b r a t e d i n conjunction with the Data AcquiSystem input  (DAS). signals  This i s to ensure that the i s such that i t s outputs 35  amplification  of  remain l o g a r i t h m i c .  DATA LINES DI09-7 FROM/TO CPU  TTTTTTTT18  11  Q  13 IR5  J_L5  3  ADC  4 7  a  &  +25 . V REF  3  20  - f  sv  1 2 10  Pl  +5V IEF  C3 H H C5  P2 cs  H H  C5  -15V  C5  +15V  15V POWER SUPPLY F i g u r e 12.  P i n diagram of the b l u e c i r c u i t board i n the C3* i s 3 C 2 - c a p a c i t o r s i n s e r i e s . Unused S/H p i n s a r e not shown. O f f s e t t r i m p o t s , R l , f o r A are not shown. (See Appendix B.1 f o r p a r t numbers.) 36  DAS.  DAS.  The  outputs  of the Pod and the Log Amp  are connected  Two  circuit-boards comprise i n the DAS.  A 15-V  to  power  the  supply  (white) board produces the necessary power f o r the components on  the  principal  12.  (blue)  board.  The l a t t e r c i r c u i t i s shown i n Figure  (Part numbers are given i n Appendix B.l.) The Pod and Log Amp operational hold  amplifiers (A) which are each followed by a  chip (S/H).  two  outputs are connected to the inputs of  used  switching device.  are grounded.  sample-and-  The outputs from the l a t t e r are then connected  channels of an 8-channel analogue multiplexer (MUX)  electronic  two  The MUX  The other s i x channels,  to  which i s that are  s e l e c t s one output of the S/Hs  an not  a f t e r the  other, passing the signal onto the next chip. The  next chip i s an 8-bit Analogue-to-Digital Converter  which i s the p r i n c i p a l component of the DAS. vert  the analogue inputs,  input  at the time,  the S/Hs  input voltages u n t i l the MUX The computer. the  I t s purpose i s to  held steady by the S/Hs,  format that the microcomputer understands.  (ADC)  to the  con-  digital  As i t can only convert  become necessary f o r maintaining s e l e c t s them f o r the  one  those  ADC.  output of the ADC goes out on a ribbon cable to the microThe  computer  cable also c a r r i e s the sequence of i n s t r u c t i o n s to  the various DAS components i n  order  to  from  properly  organize the conversion of the signals. Besides powering the As, supply  the MUX  and the S/Hs,  also operates the shutter i n the PMT.  the  +/-  15-V  The l a t t e r opens  when  the voltage i s changed from -15 to +15V,  and closes when the p o l a r i t y  is  are  reversed.  The  shutter  movements  37  co-ordinated  with  the  activation  of the DAS by a mechanical toggle s v i t c h c a l l e d Switch  A  on the front face of the DAS box. Lastly, tronics  the +5V  l o g i c l e v e l required i n CMOS  digital  [433 i s obtained by connecting the +15-V l i n e to a  regulator chip,  <P2 i n Figure 12).  elecvoltage  The +5V output also provides the  voltage drop across the Pod c i r c u i t .  I t i s also,  to another voltage regulator-reference chip,  i n turn, connected  PI to produce the 2.5V  required f o r the ADC reference. The microcomputer i s the 8-bit, t i n g system. The  I t s main advantages are i t s p a r a l l e l and s e r i a l  while  Bank #1 contains the usual transient programming area,  Bank #2 p a r t i a l l y shadows  Bank #2 only [44,453. As port,  i t . A l l ports are accessible  from  Bank #3 i s video memory.  the Osborne controls the DAS d i r e c t l y through the  parallel  considerations such as the t r a n s i t i o n time between bank had to  taken  i n t o account during programming i n order to optimize  data acquisition. this  ports.  disadvantage of t h i s machine i s that i t i s configured with three  memory banks.  be  64-K Osborne 1 with CPM opera-  the  Although the ingenuity of the e l e c t r o n i c design of  configuration  d i d not go unappreciated,  the time required  to  understand and work with i t could have been put to better use. The produces  l a s t component i n the system i s a dot-matrix the hard-copy r e s u l t s .  I t i s also  accessed  p a r a l l e l port (by Centronics communication protocol). is  printer  through the As no printout  ever required u n t i l a l l scanning has been completed,  competition f o r the port between the p r i n t e r and the DAS.  38  that  there i s no  3 .  3  O t h e r Two  philic  M a - f c e? r i a J _  contact lenses are used during scanning.  soft contact lens which i s f i t t e d f i r s t .  alleviate  the  mildly  is a  to  is  to  repeated  The second lens (Luma" lens) i s  o f f s e t the power of the cornea.  pliable plastic,  hydro-  I t s purpose  discomfort of the second lens without using  i n s t i l l a t i o n of t o p i c a l anaesthetic. plano-concave  One  It i s made  and provides the "window" f o r  of  viewing  a the  fundus. The of  s o f t lens i s attached to the cornea by the surface tension  tears and the hard lens attached to the soft lens by the  tension  of viscous methyl-cellulose.  surface  Saline s o l u t i o n was t r i e d  but  did not hold the Luma" lens i n position. I t also caused i r r i t a t i o n i n some subjects. The Their 14.  two  band-pass f i l t e r s used are Spectrotech SE4  and  SB5.  transmission p r o f i l e s and overlap are shown i n Figures 13 Their cut-off wavelengths are i n accordance with  for f l u o r e s c e i n i n water. filter only  passes  (Refer to Section 3.1.)  and  specifications  The SE4 e x c i t a t i o n  wavelengths between 453 and 493 nm  only.  the main emission peaks between 509 and 612 nm are  Similarly, transmitted  by the SB5 b a r r i e r f i l t e r . The  smaller  (blue) SE4 i s mounted i n the  also  holds the i n t e n s i t y monitoring chip.  path  of the beam before scanning.  placed shutter  between  the  I t i s inserted  The larger green f i l t e r  output of the f i b r e optic  above the PMT.  holder-slide  and  the  which  into  the  (SB5)  is  electronic  I t s large aperture ensures that a l l signals  from the conduit pass through i t before a c t i v a t i n g the 39  PMT.  40  ,T'° rrrri  F i g u r e 14.  F i l t e r Overlap. E2 and SB were used.  41  A model eye was also constructed. slit  lamp translations.  However,  I t s purpose was to c a l i b r a t e  the c a l i b r a t i o n s were made with a  more precise micrometer t r a n s l a t i o n stage f i x e d on an o p t i c a l (Refer  to Section 4.1.)  bench.  The model eye i s also used as a sample  cell  (for plasma scanning). A cross-sectional p r o f i l e i s shown i n Appendix B.3.  3.4  T h e Most  Although chosen  S o f t w a r e  programmes  were  written i n  Microsoft  (MBASIC).  a compiled version c a l l e d CBASIC was available, because i t had many b u i l t - i n functions f o r  manipulation. the DAS.  BASIC  file  MBASIC was and  string  The one exception was the programme f o r the control of  Written i n 8080 Assembly Language codes [46], i t s " l i s t i n g s "  f i l e i s c a l l e d DAS.PRN.  Certain programming "habits" were developed  because of r e s t r i c t i o n s (and economy) i n the use of memory space. For example, is  many of the MBASIC statements written were concatenated,  allowed by the language.  "Free" variables were re-used  as  wherever  possible. (Refer to Appendix A f o r a l l programme l i s t i n g s . ) S p e c i f i c subroutines are c a l l e d from menu programmes. The  first  such menu, SCANMENU.BAS d i r e c t s control to one of three subprogrammes f o r scanning and f i l i n g .  These are VITSCAN.BAS and PLASCAN.BAS which  are stored on the disk i n Drive B, itself,)  and a subroutine (in SCANMENU.BAS  c a l l e d SUBJECT DATA ENTRY.  A subroutine,  merged above SCANMENU.BAS i n the memory bank.  when c a l l e d ,  When a scan i s  is  ended,  control i s returned to the menu. Prompts  to operate the data a c q u i s i t i o n were written into t h i s 42  set  of subroutines to enhance "user-friendliness".  ponses  are indicated by h i t t i n g the ESC and ANY  tively. the  NO and YES  (other) key  res-  respec-  This association i s appropriate as the ESC key i s located at  upper l e f t corner of the keyboard and has l i t t l e probability  being mistakenly activated,  of  especially when l i g h t s are dimmed during  scanning. A  version called DAS.ASM was f i r s t prepared using the s p e c i f i -  cation  sheets  guides  [433.  of the various electronic components i n It  the  vas then assembled by the 8080 two-pass  DAS  as  assembler  provided v i t h the Osborne 1, producing the l i s t i n g s , DAS.PRN as shovn i n Appendix A . l . denote  the  language  Note that the l e f t four hexadecimal (hex)  memory  addresses  (in Banks #1 and #2)  of  codes given by the next 2 to 6 hex d i g i t s .  the  numbers machine  Entry and  exit  loops to Bank #2 are c l e a r l y marked. The port status test i s executed only on the f i r s t entry where, if  necessary,  the port-controlling.  MC6821, i s re-configured to suit the  Peripheral Interface DAS.  The strategy of t h i s subroutine i s simple. port,  p o l l Switch A u n t i l the toggle i s up.  t h e i r inputs simultaneously, is  not busy.  Next,  then hold them.  order the MUX  Adaptor,  After preparing the  T e l l the S/Hs  to sample  Ascertain that the  ADC  to switch on and the ADC to begin  d i g i t i z i n g the pod-S/H output. When the conversion i s done, store the r e s u l t at a s p e c i f i c address (D1D2 hex) i n memory. is  ready.  Now  Amp-S/H output. interrupt  order the MUX This time,  Make sure the  and ADC to do the same f o r the store the answer at D1D4  hex.  status l i n e (Switch A) and put the r e s u l t at D1D0 43  ADC  R/M-Log Test hex.  the Go  back to the MBASIC c a l l i n g subroutine. When hex  for  SCANMENU.BAS i s loaded,  the machine language code numbers (from DAS.PRN),  abovementioned are  loaded  They  i t reserves the area above  are  r e s u l t s after each c a l l .  placed  into the reserved  memory  by  the  VITSCAN.BAS  immediately  is  loaded,  codes  is  called.  hex.  a checklist of the VF  displayed and the f i r s t c a l l to DAS.PRN  the  DATA-READ-POKE  sequence of commands, starting at memory address D1D6 If  and  The machine language  only i f one of the f i r s t two subprogrammes  D1CF  system  is  is  immediately  made; usually to measure the various intra-ocular distances.  VITSCAN  . BAS marks a position by sounding a "beep" when any key i s depressed. When  "landmarking"  positions  i s ended,  i t displays  the  corresponding to consecutive beeps,  difference  between  then asks whether  to  a prompt to ascertain the eye to  be  repeat landmarking or continue on to scanning. Before scanned left.  scanning  i s given:  A  is  of the  (other) key  for  the  a set of "SCANNING INSTRUCTIONS" i s displayed. 3-second  be-varied by software to s u i t the time  loop.  constant  DAS.) The  the  ANY  toggled up to i n i t i a l i z e the DAS i n a  (This delay time may  in  ESC for the right eye;  Once answered,  Switch  begins,  d i g i t i z e d outputs of the Pod and the Log Amp  are displayed  (approximately) the f i r s t second of t h i s delay loop,  after which,  screen i s blanked.  The remaining time of the loop i s f o r  adaptation" by the subject and the system to reduce the noise up by the  "dark picked  PMT.  A "beep" sounds to mark the position of the retina (where every  44  scan  must  Control  begin),  as  well as to cue the user to  begin  toggles back and forth between Banks #1 and #2.  return  from  PEEKed.  The  DAS.PRN, data  the abovementioned  specified  scanning. After  each  addresses  are  are transferred to elements of two 2x1600  arrays  (depending on the eye being scanned). Switch A i s tested to ascertain that the scan i s to continue or to stop. The  maximum time available f o r scanning before the arrays  f i l l e d i s approximately  25s.  matically  the scanning loop to f l a s h "You  exits  memory...".  from  Alternatively,  If they are used up,  Switch  A may  are  VITSCAN.BAS autoare  out  be toggled down to  end  of a  scan. Upon leaving the scanning loop, ting  and f i l i n g subroutine.  the l e f t drive;  VITSCAN.BAS goes into a  plot-  Left-eye data are f i l e d on the disk  in  right-eye on the right. In t h i s way,  a one-diskette-  per-subject-eye system of data storage i s maintained.  This f l e x i b l i t y  allows f o r both eyes of one subject to be tested, or, two subjects to be  examined within the same period by assigning one diskette to  subject-eye. examined,  (As  many  subjects as time r e s t r i c t i o n s allow  but only the diskette i n the LOGGED drive may  may  one be  be changed -  a quirk of MBASIC.) An and  interrupt i s included to enable an "ABORT" during  filing.  confirm  It i s activated by depressing any key.  the "ABORT" appear.  prompt to continue scanning. position  An p o s i t i v e response  plotting  Two prompts produces  to  another  A negative reply returns control to the  within the plotting and f i l i n g subroutine where the  rupt was activated.  45  inter-  The  temporary  f i l e that stores the data i s renamed  when  the  p l o t t i n g and f i l i n g subroutine has been completed and i t i s confirmed that  the  data  i s to be saved.  response i s a f f i r m a t i v e ; written example,  or  The filename i s  entered  when  the  otherwise, the temporary data f i l e i s over-  erased l a t e r .  F i l e s are named by  the  p.i.  time.  For  the 3-minute p . i . bolus scan i s named 3.DAT where the f i l e -  type, ".DAT", i s automatically juxtaposed to the "3" entered. The prompt to continue scanning appears next. I f scanning i s to be  continued,  there i s a choice to begin again at SYSTEMS CHECKS or  at SCANNING INSTRUCTIONS. If the scanning mode i s to be  terminated,  c o n t r o l i s returned to SCANMENU.BAS. In  the  plasma scanning option,  s i m i l a r c a l l s to DAS.PRN  are  made.  No p l o t t i n g loop i s required i n t h i s case as the s l i t lamp  fixed  i n position.  two  (See Section 4.3.)  arrays of 1000 elements.  The arrays used are  is  smaller:  An averaging subroutine i s immediately  entered when the scanning loop i s e x i t e d . A s i m i l a r i n t e r r u p t capabil i t y i s also i n s t a l l e d . The displayed.  results The  of the averaging and the standard deviations  time of the blood sampling i n minutes p . i .  are  is  then  entered. Three 55-element arrays are used to hold the r e s u l t s of each plasma sample scanned.  At the end of scanning,  the arrays are f i l e d  i n PLASMA.DAT i n the appropriate subject-eye d i s k e t t e . The VITSCAN.BAS  SUBJECT DATA  ENTRY subroutine i s  usually  and PLASCAN.BAS have been executed.  used to enter pertinent subject i n f o r m a t i o n :  called  after •  This subroutine  is  name, age, eye that was  scanned, date, lengths of i n t r a - o c u l a r distances recorded by both the 46  slit  lamp and ultra-sound scans,  comments, on  volume of dye that  was  injected,  and observations. The l a s t category i s used to enter notes  a p a r t i c u l a r scan or the subject's medical history.  A l l informa-  tion entered i s f i l e d i n SUBJECT.DAT i n the appropriate diskette. At the end of an examination period, there should be a .DAT  file,  a  PLASMA.DAT f i l e ,  subject-eye diskette.  and the scanning .DAT  SUBJECT  files  on  a  A l l other non-.DAT f i l e s are erased. For l a t e r  analysis, a working copy of the data diskette i s always made. The execution  second of  menu  routine  i s BATCHRUN.BAS.  individual subprogrammes,  It  allows  the  or a s p e c i f i e d sequence  of  subprogrammes. On activation, a menu shows a l l the subprogrammes with b r i e f descriptions of their purposes. to  be run i s entered and confirmed.  The order of the subprogrammes The f i l e directory of the  data  diskette next appears, and the f i l e s to be analyzed are then entered. After execution  printer  status  and paper  supply  have  been  checked,  i s begun by CHAIN MERGEing the f i r s t subprogramme i n the  sequence. Control i s then transferred to the task. On completion, the next subprogramme i n the sequence i s loaded over the f i r s t , and so on until  the s e r i e s i s completed.  The following describes the  subpro-  grammes and t h e i r tasks. REDUCE.BAS the to  averages the raw data i n each . DAT  file,  reducing  number of data-points from a possible maximum of 2x1600 per f i l e a possible averaged maximum of 3x256 - Pod output,  output  and  i t s standard deviation.  because of memory space r e s t r i c t i o n s , flowchart i s shown i n Figure 15.  47  The  algorithm,  averaged although  R/M slow  i s not d i f f i c u l t to follow. A  X0=Pod starting position; XX=final X=X-X0; YY=Averaged Y; Z2=YY's S.D. at XX; U=YY at previous X, i . e . X - l ; T=chosen Y to average about; If X=X- ", set U=T; L=averaging i n t e r v a l about T.  jRead in .DAT f i l e : X0; <X,Y) pairs; f i n d X- " and X"*". 4  1  Store XX,YY, ZZ i n .AVG files.  •j STOP )  Yes  ©  t£) Bin a l l Y's of t h i s X.  Choose the Y that occurs most times and i s closest to U. Set t h i s Y=T.  Find the Scatter of these Y's. No  Yes  No  L=all Y's. Yes  ©4  Set L=[T-5, T*5],  T >U Yes No  Set L=CU,T+5], Set L=CT-5, U3.  Average over a l l the Y's in L to f i n d YY and ZZ, and XX=X-X0. Set U=YY.  Figure 15.  FLOWCHART f o r REDUCE.BAS.  48  Basically,  REDUCE.BAS considers the scatter of the R/M outputs  at each Pod output. are  included  result  I t sets an i n t e r v a l within which a l l R/M outputs  i n the averaging.  This i n t e r v a l i s dependent  of the averaging of the Pod output before  on the  i t . Hence, the  averaging process i s carried out point by point i n ascending order of Pod output from the posterior vitreous to the cornea. The r e s u l t s are zeroed  by  the r e t i n a l position (at the "beep")  and  converted  to  concentration units before being stored i n a .AVG f i l e with the same p.i.  time filename. After  •DAT  a resultant .AVG f i l e has been entered,  f i l e i s destroyed.  the associated  This action frees diskette space f o r  later  f i l e s . This i s the reason for creating a working copy of the o r i g i n a l data diskette. This technique of data reduction or "smoothing" was chosen over the  usual  scanning.  methods Also,  such as c u r v e - f i t t i n g because of the method of  as  mentioned  before,  t h i s VF system i s a  averaging DAS as opposed to the "spot" system [383. biased",  time-  I t i s "backward-  as i t i s dependent on the average of the previous Pod posi-  tion because of the capacitive time constant (approximately 2.2s)  in  the electronics. Several background scans are usually made. i d e n t i f i e d as "0",  "00",  They are .DAT f i l e s  "000", etc. B/G.BAS contains a background-  f i l e - t e s t i n g loop to ascertain that multiple pre-injection scans were made.  I f the loop f i n d s several such scans,  B/G.BAS  takes  these  already averaged background scans, aligns (by the location of r e t i n a l position)  and  averages over a l l of them to produce one f i n a l 49  0.AVG  file.  A l l other background .DAT and .AVG f i l e s are then deleted  free diskette space.  If only one background scan was  made,  to  B/G.BAS  returns to BATCHRUN.BAS to the next subprogramme i n the series. MINUS.BAS  subtracts the f i n a l 0. AVG f i l e from a l l other  .AVG  f i l e s . Two sets of results are produced by two types of alignment: a) by matching the positions of the retina as they were marked on each scan  (at the beep); or,  b) by searching out and matching the CR peaks. Subtraction i s then carried out after each alignment.  The results by  (a) and (b) are f i l e d i n .RET and . CRP f i l e s respectively.  Again the  filenames are the p . i . times of scan. SUDATA.BAS  prints  out the subject  and  scanning  of the information i n the SUBJECT.DAT  data.  A  file  is  formatted  output  produced.  The F-numbers (in Section 2.2) are calculated and printed.  The routine reads through every .AVG, the  concentrations  These  .RET and .CRP f i l e ,  at a set of s p e c i f i e d points  points are the retina,  the CR,  i n each  compiling profile.  lens and corneal peaks, the  minima and the centres of the vitreous and the aqueous chambers, points (that are i n the vitreous chamber only) which are 0, 12,  and  3, 6, 9,  15... mm from the retina. At the end of compilation, SUDATA.BAS  calculates,  when appropriate,  the performance s p e c i f i c a t i o n s : LLoD,  AR and R. BLOOD.BAS i s the c u r v e - f i t t i n g subprogramme that calculates the areas under the plasma profile. converted buffer  PLASMA. DAT i s read, and the data are  to concentration units with a constant that  dilution  and  the unbound 50  fluorescein  factor  accounts f o r - 17'/.. (See  Section 3.1 and C47],) To fuging,  calculate t h i s constant,  i t i s assumed that after  centri-  a l l the red blood c e l l s have been removed to one end of the  haematocrit  but that  throughout  the  the fluorescein i s homogeneously  plasma.  t r a t i o n at sampling time, the BRB at a l l times,  distributed  I f c* (t) i s the plasma fluorescein t,  concen-  and only 17% i s available to penetrate  the area under the plasma-fluorescein  profile  i s given by  (23)  I(t) =  0.17  C"(T)  dT  T • *  If X ml of each spun sample i s diluted with Y ml of buffer,  then the  resultant concentration, c»(t), measured by the s l i t lamp i s given by  (29)  c» <t) * X • c» <t) / { X • Y )  .  Eq. 28 becomes  f>T - «  (30)  I(t) =  0.17 { 1 • Y/X }  C (T) P  dT  T • •  The factor before the integral depends on the amount of free fluorescein (17%),  and the volumes of plasma sample and the buffer used i n  the d i l u t i o n process. Note that any amounts of X and Y may be used. A i.e.,  set  of four polynomials of order 2 i s f i t t e d to  the  (2+1)-parameter c u r v e - f i t t i n g [34]. The polynomials are: 0)  y  =  A  +  B»x  +  1)  y  =  A  •  B * log x 51  C » x* +  ;  C » ( log x )•  ;  data,  They  2)  y  =  A * exp { B * x  •  and, 3)  y  =  A  C/x*  were  blood.  chosen  •  B/x  +  C » x* }  ;  to approximate the removal of the  dye  from  the  The d i f f i c u l t y i n producing a f a s t and e f f i c i e n t algorithm i n  a non-linear f i t to the sum of two negative exponentials (expected of the  f a s t and slow decays of the two-compartment model  Section  mentioned  in  2.2 and [28]) forced such a method f o r estimating the plasma  integral.  I t i s also d i f f i c u l t to j u s t i f y such a non-linear f i t to a  few data-points only. (Refer to Sections 4.2 Areas  under  integration  4.3.)  and  - (1)  the best f i t are calulated as follows  lower l i m i t i s t=0. 5 minutes p . i .  (2) The upper  the limits (3)  are the measurement scans' p . i . times (given by the filenames). The  area  right-angle the  best  0 to 0.5  from  minutes i s approximated by the  area  of  triangle with base 0.5 and height equal to the value f i t at 0.5.  superimposed  on  (Refer to Section 2.2.)  the raw data are produced.  Plots with  the  a of  fits  The goodness of f i t i s  determined by the reduced chi-square of each f i t . At  the end of the four f i t s ,  chi-square i s chosen. FIT f i l e ,  the one with the lowest  reduced  Its code number (0-3) i s entered i n a  PLASMA.  followed by the c o e f f i c i e n t s (A, B and C) and their calcu-  lated errors,  the upper limits of each integration,  the areas up to  those l i m i t s , and their errors. C/VAZ.BAS  follows closely the algorithm of Section  subprogramme uses .RET all  profiles  retina  as  and .CRP  f i l e s only. When .RET  52  each  scan.  This  f i l e s are used,  are aligned for subtraction by the positions  marked at the beginning of  2.3.  of  Similarly,-  the .CRP  f i l e s are aligned by the CR peaks before any subtraction. The bolus .RET f i l e and a measurement .RET f i l e are read taneously.  The  permeability indices,  penetration ratios, lated  P  1  (of Section 2.2)  simul-  and the  PR3 (of Section 2.3), as well as PR3» are calcu-  f o r each p r o f i l e .  The CR correction condition,  the left-hand  side of Eq. 6, i s also calculated. Two  types  of bolus  correction  are undertaken.  The  involves  subtraction of the bolus from the measurement scan  applying  the CR bolus r a t i o .  1  the  subtraction  the  CR bolus ratio.  calculated.  without  (Refer to Section 2.5.) A plot of the  r e s u l t s between 0 and 11 mm from the r e t i n a i s produced, calculations of P ,  first  followed by  PR3 and PR3*. For the second type pf correction,  i s repeated but with the bolus p r o f i l e modified A plot i s produced,  and the three  by  parameters  The procedure i s carried out f o r the . CRP f i l e s with the  same p . i . times. If more than 10 non-negative points e x i s t within the posterior half of the vitreous chamber after each subtraction,  C/VAZ.BAS f i l e s  them i n .CV# f i l e s on the data diskette. The f i l e s are named with the same p . i . time filenames,  but, with the following f i l e t y p e s :  .CV1  'unaltered bolus subtraction of .RET f i l e s " ;  . CV2  "modified bolus subtraction of .RET f i l e s " ;  . CV3  "unaltered bolus subtraction of .CRP f i l e s " ;  . CV4  "modified bolus subtraction of .CRP f i l e s " .  If many measurement scans are made, out of storage space. C/VAZ.BAS  the data diskette can run  To prevent t h i s from halting the calculations,  contains an error-trapping subroutine 53  which  automatically  files  the .CV# f i l e s on the diskette which contains the MBASIC  grammes i n the other drive. . BAS,  some  programme  By studying  the plots produced by C/VAZ  of the . CV# f i l e s can be deleted.  diskette  can then  pro-  The .CV# f i l e s on the  be transferred  back  onto  the data  diskette. SLOPES.BAS  checks the r e s u l t s of C/VAZ.BAS by c u r v e - f i t t i n g at  the . AVG-file l e v e l , as  those  using the same set of (2+1)-parameter functions  i n BLOOD.BAS.  curve-fitting.  A  Only vitreous data are considered  vitreous  p r o f i l e i s divided into three  i n the sections:  posterior, mid-, and anterior vitreous. The four functions are f i t t e d to one section at a time. i s chosen to represent When  The f i t v i t h the lowest reduced chi-square  that section i n a l l l a t e r calculations.  a l l the best f i t s have been calculated f o r each  section,  SLOPES.BAS uses them to estimate the following: a) the concentrations  after background subtraction, c ( x , t ) .  b) the penetration r a t i o s , PR3», before and after bolus subtraction. c) the gradients, c ( x , t ) . 9  d) the d i f f u s i o n constants, The  D.  values chosen for x are 3 and 9 mm from the retina,  as well  as  3mm from the posterior surface of the lens. PR3* i s found at the 3-mm point  from  the retina and the lens without  CR  bolus  correction.  Alignment i s by RET only. D i s approximated by converting into  a  difference equation [34,48].  c a l c u l a t i o n s at the 3-mm  and 9-mm  the d i f f u s i o n equation,  This approximation uses the  points. I t i s  54  Eq. 1, .  (31)  c<9,t)/t  t i s i n minutes p . i .  (32) The  D  =  =  D • f c ( 9 , t ) - c'(3,t) )/< 9 - 3 ) . 8  Solving f o r D,  c<9,t) » t-' * 10- / { c ( 9 , t ) - c ( 3 , t ) } 1  8  em's".  a  10- factor appears when converting to centimetres and  seconds.  1  These  values  LUND.BAS  of PR3* and D may be used as  curve-fitting  as  initial  estimates f o r  well as checking the solutions  of RET  C/VAZ.BAS. PLOT.BAS  i s an  e a r l i e r version of  i n d i v i d u a l scaled-down plots of any .AVG, information  gathered  SUDATA.BAS.  I t produces  .RET or .CRP f i l e . A l l the  by SUDATA.BAS i s also  collected  and  printed  below each plot. Comparison of data and plots by inspection i s d i f f i cult. DRAW.BAS i s the plotting subprogramme that .BAS. log  complements  SUDATA  This subprogramme uses the f i r s t entered f i l e to produce semi-  plots recorded data-point by data-point. A l l subsequent plots are  scaled and superimposed on the f i r s t plot. I t s p r i n c i p a l advantage i s that as  i t can superimpose the p r o f i l e s from d i f f e r e n t subjects as well producing  "fully  stretched" p r o f i l e s .  c l a r i t y when too many f i l e s are superimposed, means  Limited by the loss  of  the routine provides a  f o r easy comparisons of the p r o f i l e s from d i f f e r e n t  subjects,  and the evolution of fluorescein p r o f i l e of one subject. The stated  l a s t routine written was LUND.BAS.  i n Section 2.4.  The algorithm i s that .  The routine - largely translated  from the  FORTRAN subroutines i n Bevington [34] - uses the Marquardt  gradient-  expansion method of non-linear c u r v e - f i t t i n g .  55  Although i t i s able to  read data from any f i l e ( - t y p e ) ,  LUND.BAS i s usually run on data from  .CV# f i l e s . The following are required at each run: a) chosen f i l e from which data are read. b) i n t e r v a l within which data are to be f i t t e d . c) i n i t i a l estimate of P, and i t s step increment. d) i n i t i a l estimate of D, and i t s step increment. The retina.  chosen Initial  results  i n t e r v a l i s usually between 1.5 and 6 mm from P  and D values depend on the expectations  f o r the particular subject i . e . ,  and SLOPES. BAS. The  the r e s u l t s of  the  of the C/VAZ.BAS  Increments are usually 107. of the i n i t i a l estimates.  subprogramme tests the f i t by c a l c u l a t i n g  a reduced c h i -  square rather than the S* of Eq. 23 i n Section 2.4.  The weights are  the S.D.  of the data as calculated by REDUCE. BAS (or others),  at each Pod position.  i.e.,  I t i s also possible to use the LLoD ( i f known)  i n Eq. 24. The reduced chi-square f o r the i n i t i a l values entered i s found. The subprogramme then goes on to f i n d  better  f i t s by varying P and  D by the increments entered. After each f i t , a new reduced chi-square i s found.  I f i t i s smaller,  the subprogramme can be terminated.  If  not,  i t i s used to determine the d i r e c t i o n to vary D and P f o r the  next  f i t . The  hardware zero),  search  interrupt  can only be terminated i n three  (abort) or,  (2) overflow  errors  ways  - (1)  (division-by-  (3) reduced chi-square less than 2. As poor i n i t i a l estimates  may r e s u l t i n slow convergence and long computing time, t h i s subprogramme i s only used i n overnight runs.  56  X V .  C A L I B  In  F£  t h i s chapter,  A T O M S  A N D  P R O T O C O L  the c a l i b r a t i o n s of various components of the  VF system and the preparation of subjects f o r scanning are described. The r e s u l t s of the former are given i n Appendix C. As the ADC i s an 8-bit converter f o r a maximum input of 5V, one d i g i t increment of i t s output 256 "Osborne/DAS" units - 0 to 255 - i s a *19-mV change of input voltage. that are expected during a scan. must  A l l c a l i b r a t i o n s must be i n ranges The gains at every stage i n the DAS  be c a r e f u l l y adjusted so that the f u l l range of the ADC can be  used. In t h i s way, the s e n s i t i v i t y of the VF system i s optimized.  •4. 1  InstruTTient  Preparation  The operational amplifier (A) receiving the Pod output, and the S/H were  o f f s e t and zeroed.  The gain was adjusted to approximately 12.  This allowed a net displacement of the s l i t lamp of approximately mm.  The focus  displaced  of the probe i n the vitreous  chamber  22  however, was  more than t h i s value because of the F-numbers (in Section  2. 2). The the  slit  micrometer  Pod (or s l i t lamp translation) was calibrated by placing lamp microscope assembly on an o p t i c a l bench translation stage was fixed.  on which  a  The micrometer was used to  displace the s l i t lamp i n precise, 0.025-in increments. The r e s u l t of each  displacement was recorded from the computer monitor.  least-squares slope  f i t was carried out on the data collected.  A  Only the  of the f i t i s required as the zero position changes with  57  linear  each  scan. (See Appendix C.1.) As input  the Log Amp and the DAS are both capable of amplifying  any  signal their gains have to be adjusted i n tandem allowing  low  pre-injection allowing  scan signals to be d i f f e r e n t i a t e d from noise  but not  high concentration signals to be amplified to greater  than  the maximum 5V that the ADC can convert. The  first  step was to ascertain the range over which the Log  Amp remained logarithmic.  At low voltage inputs, the Log Amp did not  maintain i t s function, because i t was performing close to "zero* at 1 mV (the logarithm  of 0 was of course not defined).  For input values  up to 10V, the Log Amp performed s a t i s f a c t o r i l y . Other conditions were: a) R/M set at 0-exponent outputs a maximum signal of about 10V for a maximum concentration of about 100000" ; 1  b) Log Amp output i n the range of 0 to 5V f o r the ADC using 1 mV to 10 V as the input range from the R/M to the Log Amp. The Log Amp was adjusted so that i t s output was logarithmic i n the  range  of input voltages.  (See Appendix C.2. )  The  operational  amplifier to which the output of the Log Amp was connected,  was then  adjusted to a gain of 1. Samples  of various concentrations of sodium fluorescein  were  freshly made at each calibration because solutions lose t h e i r fluorescence e f f i c i e n c y over long storage periods. As the pH of the solvent affects  i t s efficiency,  the appropriate buffer solution  (pH  7.4,  discussed i n Section 3.1) was also prepared. A pH meter was calibrated at 20C and pH 7.0 using a standard pH 58  7.0 solution.  The Sorensen's Phosphate buffer solution at pH 7.4  was  then made up as follows a) Prepare a 0.9077. weight by volume (or, 0.067 moles) solution of monobasic  potassium phosphate i n demineralized  Solution  distilled  water.  "A".  b) Prepare a 2.397. weight by volume (or, 0.067 moles) solution of sodium phosphate i n demineralized d i s t i l l e d water. - Solution  "B".  c) To get 100 ml of a pH 7.4 buffer solution, f o r example, mix 19.7ml of A and 80.3ml of B, i.e.,  mix i n A:B volume r a t i o of 1:4.076.  As an alternative, once the pH meter was calibrated, the pH 7.0 buffer concentrate, (the standard used to c a l i b r a t e the pH meter) was used as a "base" to make the pH 7.4 buffer. because  This method was possible  the concentrate was made from sodium  phosphate.  Hence,  by  adding potassium phosphate to the solution at pH 7.0, a pH 7.4 buffer was obtained. This method was used to prepare the large volume of the pH  7.4  solution  needed  that was required f o r  f o r calibration,  the  sample  and also f o r d i l u t i n g plasma  concentrations samples  from  blood taken during scanning. Sample concentrations were made by a d i l u t i o n method. a) Weigh an empty test-tube with seal i n place. b) Add an amount of sodium fluorescein powder, then weigh again. c) F i l l the test-tube with the pH 7.4 buffer and weigh again. The other  first  samples  assuming gible  prepared were  sample was a "master solution" from derived.  I t s concentration  was  which a l l .  estimated  that the mass and volume of sodium fluorescein were  compared  to  the  mass  and 59  volume  of  added  buffer.  by  negliThe  concentration of the master solution was then  (33)  c"  where  (a),  before.  a  (b)  The  = { (b) - (a) } / { (c) - (b) }  and  mass  (c) are the r e s u l t s of  of"  1  the  stages  the dissolved s a l t s was also  mentioned  assumed  to  be  negligible. The following procedure was used to make other samples: a) Weigh an empty test-tube with i t s seal. b) Draw a small volume of c"  a  with a 1-ml  tuberculin syringe,  place i t i n the test-tube and weigh i t . c) Add buffer then weigh again. The concentration of the new (34)  c = c"' * { (b) - (c) ) / { (c) - (a) >  where (a),  (b) and  d i a t e l y above. and  solution was"  1  .  (c) are the r e s u l t s of the steps discussed  imme-  Other samples were made by varying the volumes of c"  8  the buffer i n order to prepare concentrations between 5 and 9000" .  Note  1  that  the error due to the mass  of  the  fluorescein  became less with greater d i l u t i o n . The prepared concentrations were placed i n several sample c e l l s (cuvettes)  which were clamped i n front of the s l i t lamp.  was scanned through i t s 1-cm e f f e c t s of attenuation. tration  profile  Each  cell  depth. This was carried out to study the  (See Appendix C.5.)  was used to find the  The peak i n each concen-.  R/M-Log  Amp-DAS  calibration  curve. The reason for t h i s was explained i n Figure 2. The  R/M  output increased l i n e a r l y with the concentration  60  that  i s scanned but not the output of the Log Amp. equation  was  Osborne/DAS functions  expected units  to  to  be an exponential  concentration  Hence, the c a l i b r a t i o n that  values.  used were those i n BLOOD.BAS.  translates  The  the  curve-fitting  (Refer to Section 3.4.) The  best f i t was the one with minimum reduced chi-square.  The mathemati-  c a l s o l u t i o n i s given i n Appendix C.4. Other  s l i t lamp c a l i b r a t i o n s included s l i t  dimensions,  angle  between e x c i t a t i o n beam and probe, and the i n t e n s i t y of the beam. The slit  s i z e used was 2 X 0.1 mm.  As the length did not affect the AR  ( i n Figure 2), t h i s magnitude was chosen to provide better v i s i b i l i t y when focussing at the r e t i n a . The width of the s l i t was defined by the r u l i n g s on a red blood cell  counter (or hemacytometer).  The v e r t i c a l and h o r i z o n t a l  formed a g r i d of 1 mm* with subdivisions of 1/400 mm2. focussed  on  the g r i d .  maximum  (35X)  desired s i z e .  lines  The s l i t  was  The microscope oculars were adjusted to  the  magnification  and the s l i t  width  adjusted  to  the  The slit-adjustment knobs were then secured to prevent  a c c i d e n t a l reset. Slit-lamp intensity, at  a  chip  "intensity"  output used  monitored by the s o l a r - c e l l chip, was set  voltage  of 141 +/-  1  mV which  to carry out the concentration  was  the  calibrations.  lamp All  scans had to be made at t h i s LED voltmeter reading. The output had to be  checked before and after each scan to a s c e r t a i n that the  excita-  t i o n beam did not f l u c t u a t e s i g n i f i c a n t l y during scanning. When excitation  the beam  angle  between the d i r e c t i o n s of the probe  was maximum,  the AR was at a £1  minimum.  and  the  Using  the  average depths and r e f r a c t i v e indices of the model eye schematic  eye  [30],  - Gullstrands  t h i s angle was found to be approximately  (Other investigators' instruments are usually set at 14".  16°.  [31])  The entire s l i t lamp (objectives and bulb housing) was rotated, then  lacked  assembly. subject  at  This to  8°  from the translation  latter  of  the  slit  adjustment could be changed  for  particular  overcome i r i s clipping or  axis  lamp  specular r e f l e c t i o n o f f  the  plane surface of the Luma* Lens onto the probe. To of  test the s u i t a b i l i t y of a l l instrument adjustments,  the model eye was carried out.  a scan  Different concentrations were used  in the various compartments. The p r o f i l e i s shown i n Appendix C.7.  Bulb and Focussing  FOCUS  Probe Pick-up Path Right Ocular with Fibre Optic Probe  Figure 16.  Overhead view of the angles at which the s l i t lamp was set f o r LEFT eye scans.  62  4. 2  Sub j ect, A  Preparation  subject i s given a CONSENT FORM (shown i n Appendix A.17)  to  read and sign. The attendant ophthalmologist answers additional questions the subject may have and countersigns the form. The age and the weight of the subject are noted and the l a t t e r used to calculate the amount  of sodium  fluorescein  f o r i n j e c t i o n (at  14  1  body  weight). The pupils are d i l a t e d using drops of tropicamide 1% and phenylephrine period  57., repeated a f t e r 10 minutes.  After a 30-minute  waiting  f o r the drugs to take effect, the pupil diameter i s measured.  If i t i s 7 mm or larger, dilation  the scanning procedure i s started.  i s s t i l l inadequate  If  the  f o r scanning a f t e r additional tropica-  mide and phenylephrine, the procedure i s cancelled.  lamp  The subject's seated position i s explained and t r i e d .  The  slit  and chin rest are adjusted to the subject's height to  minimize  discomfort during the scanning.  The s t e r i l i z e d soft-contact lens i s  then  then the s t e r i l i z e d  fitted  to the l e f t eye,  Luma"  Lens i s  mounted with a drop of methyl-cellulose. Each mounting i s checked f o r trapped a i r bubbles. The subject then places his/her head i n position on  the s l i t  lamp assembly with i n s t r u c t i o n to press  the forehead  against the headband. The  first  scan  f o r "landmarking"  localize  the surface positions of the  anterior  lens,  uses  unfiltered  retina,  l i g h t to  the posterior and  and posterior cornea (along the optic  axis).  The  depth or thickness of each section i s displayed and noted (in Osborne /DAS  units)  by the programme,  VITSCAN.BAS. 63  The measurements are  usually done three times to check r e p r o d u c i b i l i t y . The retina i s marked by two methods. smaller, the  0.1mm  effects  Another  s l i t on the larger,  One way i s to centre  0.45mm diameter probe.  of scattering (halation) makes t h i s  However,  method  difficult.  method consists of placing the l e f t edge of the s l i t on  r i g h t edge of the probe;  the  the  then the r i g h t edge of the s l i t on the l e f t  edge of the probe. The two results are l a t e r averaged to "define" the retinal  position.  Both  methods are used within the three  measure-  ments.  (a) Retinal landmarking Figure 17.  Other For  example,  at the posterior surface of the lens,  i s seen  within  the lens tissue, and  smaller  L o c a l i z i n g techniques.  surfaces are marked by the contrast at  cation  lens  (b) Other interfaces  as there i s autofluorescence  interfaces.  a sharp  demar-  scattering  from  and r e f l e c t i o n at the interface between the  the "dark" vitreous volume.  depth  and  the  (Refer to Section  of the anterior chamber makes the measuring  64  2.1.) of  The that  compartment more d i f f i c u l t . The pre-injection scan(s) are done immediately after the length measurements retina,  are  completed.  The  started at the "beep", studied  Switch A i s toggled up.  and at the end of each scan,  When removed  background  and  the  is  i s completed,  injected and photographs  are taken. advises  (nausea),  scanning  the  Up  to  Luma* room  of both maculae  The injection i s given by a registered  the  Scanning i s  four  3.4.)  the subject brought to the photography  fluorescein  also  on  the visual plot  f o r signs of clipping or subject movement.  background scans can be made. (Refer to Section  discs  focussed  and s l i t intensity i s checked. The blue excitation f i l t e r i s  s l i d into the path of the beam;  is  probe i s f i r s t  subject of the dye's possible  lens where and  is the  optic  nurse  immediate  who  effects  and delayed effects such as yellow-coloured urine. The dye  i s injected quickly. half-emptied.  Two stopwatches  The photography  are started when the syringe i s  i s usually completed  55 to 70 s  after  the i n j e c t i o n . The  bolus scan i s the c r i t i c a l  scan as there i s a maximum time  i n t e r v a l within which i t must be made. t i o n 2.3.) events  The protocol requires a 3-minute p . i .  such  scan.  A  scans  are  (Refer to the theory i n  bolus scan. However,  as instrument intensity re-adjustments may  bolus i s never accepted past 7 minutes p . i .  fluorescein,  delay  Other  taken to study the early changes i n the l e v e l s of and,  Sec-  therefore, CR bolus t a i l i n g s . Plotting and  this  "bolus" plasma filing  usually take up to 90 s. Immediately  after bolus scanning 65  and the removal of the  Luma*  lens, a blood sample i s collected. A finger i s pricked with a s t e r i l e lancet  and  rinized  approximately  0.1 ml of blood i s c o l l e c t e d i n a  c a p i l l a r y tube (haematocrit).  The p . i .  time i s noted  hepawhen  half the tube i s f i l l e d . Another sampling i s done at about 15 minutes p. i . The  first  measurement scan i s taken around  30  minutes p . i .  Other scans can also be taken f o r r e p r o d u c i b i l i t y studies. blood  sample  procedure removed  i s taken a f t e r the Luma Lens has been  i s repeated around 60 minutes p . i . as  The t h i r d  removed.  This  The s o f t lens i s also  scanning with the s l i t lamp i s ended  after  the 1-hour  measurement scan(s). The l a s t parts of the examination are monochromatic nerve-fibre photography  and ultrasound measurements of  The  i s used to compare against the distances of the various  latter  intra-ocular  distances.  segments of the eye measured by the s l i t lamp. The ultrasound r e s u l t s are also used to scale the p r o f i l e s collected,  and to calculate  the  F-numbers f o r the optics i n t h i s VF system.  START Read and sign CONSENT Form. Answer questions. -30 min.p.i. D i l a t e pupils with tropicamide. -5 min.p.i. Mount lenses. Measure intra-ocular distances. Background scanning. 0 min.p.i. Injection and photography. 3-7 min.p.i. Bolus scanning and blood sampling. 15 min.p.i. Blood sampling. 28-33 min.p.i. Measurement scanning and blood sampling. 55-65 min.p.i. Measurement scanning and blood sampling. Nerve-fibre photography; Ultrasound scanning. Plasma-fluorescein scanning. Figure 18.  Flowchart of the scanning 66  procedure.  4.3  B l o o d - P l a s m a The  four  P r e p a r a t i o n  blood samples c o l l e c t e d are centrifuged  after the ultra-sound measurements have been made. at  2000 rpm f o r approximately 12 minutes.  the  red blood c e l l s .  immediately  The speed i s set  This adequately separates  2 ml of pH 7.4 buffer solution are measured out by a volumetric pipette cell  and placed i n each of f i v e (cuvette) dry sample  i s used as the background sample.  cells.  One  0.01 ml are dravn from  each  haematocrit using a 0.025-ml pipette and emptied into each c e l l . constant outside the i n t e g r a l i n Eq.  The  30 i s then 34.17. The c e l l s are  then scanned by the s l i t lamp. Each its such  c e l l i s placed so that no l i g h t i s r e f l e c t e d o f f  surfaces onto the probe. a  way  that  The probe i s positioned  i t i s not focussed on a i r bubbles  surfaces of the c e l l s .  any  of  visually,  in  on  the  inner  The focus of the probe i s s l i g h t l y behind the  inner front wall of each c e l l . This i s due to possible attenuation of the  incident beam owing to the high concentrations. Also, from Figure  2b, the probe must be t o t a l l y included i n the sample volume. The  f i r s t sample to be scanned i s always the background  cell.  It i s scanned once only. Samples with fluorescein can be scanned more than  once by varying the position of the probe's focus s l i g h t l y each  time. Also, plasma fluorescein samples are usually scanned i n chronol o g i c a l order of t h e i r p . i . times. SUBJECT DATA ENTRY i s usually executed l a s t .  67  V  -  A N A L Y S I S  5.1  A N D  D I S C U S S I O N  C l a s s i f i c a t i o n s The procedure vas completed on the following sample population:  TYPE  NUMBER  MALE Diabetic FEMALE Diabetic  10+2 5  MALE Normal MALE MS FEMALE MS TOTAL  Two  male  AGE RANGE 21 - 64 18 - 57  4  27 - 38  3 13  32 - 39 20 - 59  35*2  Table 2.  Distribution of subjects.  diabetic  subjects were recalled at one  and  months after t h e i r f i r s t scans to study the r e p r o d u c i b i l i t y .  three  The two  sets of r e s u l t s f o r these two subjects were averaged i n the following analysis. Only three blood samples were taken from the f i r s t two male and  the f i r s t two female diabetic subjects on whom the procedure was  carried out. Whenever  possible  each parameter studied was  tested on the basis of disease (MS or diabetes),  classified  and  age, and sex. Only  l e f t eyes were scanned because of the time r e s t r i c t i o n for the bolus, scans i . e . , a one-eye design [49], Diabetics into  with  non-proliferative retinopathy  were  separated  three groups according to the severity of c a p i l l a r y disease  68  as  shown  by fundus photography and fluorescein angiography.  consisted of subjects with no or early DRP (0 t o 5  Group D(l)  microaneurysms);  group D(2), subjects with mild to moderate non-proliferative DRP, and group D(3), subjects with severe non-proliferative DRP.  DISEASE STATES  NUMBER  D(l) - Diabetic with zero to early DRP ( 0 - 5 microaneurysms).  7M IF  D(2) - Diabetic with mild to moderate DRP.  IM 2F  DO) - Diabetic with severe DRP.  2M 2F  TOTAL DIABETIC SUBJECTS AVAILABLE  10M:5F  MSU) : stable recovering from relapse  2M:3F 1M:2F  MS(2) : slowly progressive i n relapse  0M:4F 0M-.4F  TOTAL NUMBER of MS subjects AVAILABLE  3M:13F  MS(3) : benign relapsing-remitting  1M:1F 1M-.6F  MSU) : relapsing-progressive chronic progressive  1M:5F 0M:1F  TOTAL NUMBER of MS subjects AVAILABLE  3M-.13F  MS(5) : no p e r i p h l e b i t i s  2M:5F  MS(6) : active p e r i p h l e b i t i s  0M:2F  MS(7) : i n a c t i v e p e r i p h l e b i t i s  1M:3F  TOTAL NUMBER of MS subjects AVAILABLE  3M:10F  Table 3.  Detail subject c l a s s i f i c a t i o n s .  69  MS subjects were grouped f o r a n a l y s i s i n three ways. was by the a c t i v i t y of MS at the time of the scans. comprised a  The f i r s t  The subdivisions  MS<1) - subjects who were e i t h e r s t a b l e or recovering from  relapse or had recovered completely from a recent relapse  time  of  scanning  progression activity were  MS(2) - subjects who were  or i n relapse.  either  The second was by the standard  categories of MS as reported i n case records.  again  included  and  subdivided i n t o two sections - MS(3)  and  at  the  in  slow  clinical  The subjects MS(4).  those c l a s s i f i e d as benign (only one attack) or  MS(3)  relapsing-  r e m i t t i n g (almost complete recovery from each a t t a c k ) . MS(4) subjects were c l a s s i f i e d as relapsing-progressive, third  grouping  presence  or chronic-progressive. The  separated the subjects according to the  of e i t h e r a c t i v e or i n a c t i v e r e t i n a l  were c a l l e d MS(5),  absence  periphlebitis.  or  These  MS(6) and MS(7) r e s p e c t i v e l y . As not a l l subjects  were examined f o r these states, the sample s i z e f o r t h i s grouping was reduced.  5 . 2  F - N u m b e r s  The F-numbers i n Section 2.2, were c a l c u l a t e d by SUDATA.BAS f o r each subject. intra-ocular  The average and standard d e v i a t i o n were found f o r each compartment and tested against the averages from  strands emmetropic model eye i n Table 1. using the data from 34 subjects only.  The r e s u l t s were c a l c u l a t e d  Three subjects (from Table  were excluded because they did not have the ultra-sound scans. numbers  are  c h a r a c t e r i s t i c s of the system,  separated i n t o age, sex or disease s t a t e s . 70  Gull-  the subjects  were  3)  As Fnot  MEDIUM  AVERAGE  Aqueous Lens Vitreous  Table 4.  Since  34  S.D. 0. 359 0.157 0.089  1.735 1.511 .; 1.245  Average F-numbers and their S.D.  measurements were made,  the mean values i n Table  4  were tested against those i n Table 1 using a normal(0,1) d i s t r i b u t i o n test.  A  that  the  P=l% l e v e l of significance F-numbers  was imposed.  from Table 1 were not  The results showed  applicable  to  this  VF  system: the F-numbers i n the two tables were s i g n i f i c a n t l y d i f f e r e n t . Hence, lens  the  Luma" contact lens (which replaces the Goldmann  i n Lund-Andersen's calculations)  changes the F-numbers s i g n i f i -  cantly.  2.4-  Legend  2.2A  2-1  a  •  VITREOUS  X  LENS  a  AQUEOUS  ce CD  I  u.  8*  1.6-1  CP  a  1.2-  20  SO AGE  Figure 19.  40  SO  60  70  in y e a r s  Plot of F-number results. 71  contact  5 . 3  I n t r a - o c u l a r The  lens  L e n g t h s  averages for the vitreous and aqueous chamber depths, the  thickness and the t o t a l a x i a l length as measured by  f o r male and female subjects are shown below.  ultrasound  The sample sizes  were  17 i n each group.  AGE RANGE  SEX  16-20 21-30  F  31-40 41-50 51-60 61-65  VITREOUS 15.76 17.96 16,24 16. 44 15. 94 16.13 14.45 14. 78  H  F M F F F M  LENS 3. 38 3. 67 3.38 3.81 3.96 3. 84 4.32 4. 54  AQUEOUS 4.45 3. 60 3. 96 3.56 3.42 3.48 3.22 3. 06  TOTAL  Table 5.  NUMBER  23.59 25.23 23.58 23. 81 23. 31 23. 46 21.99 22. 38  3 7 6 9 2 2 2 1 34  23.74  Average lengths of the i n t r a - o c u l a r media (in mm).  Although the states  AXIAL  sample sizes were small,  were not taken into account,  and the various disease  the above table suggests either  that a x i a l length decreases with age, or, that the lens thickens with age. The  [50]  Linear least-squares f i t s were done on each  n u l l hypothesis,  H*,  i n each case was  treatment.  that the slope  of the  straight l i n e was zero. The alternative hypothesis, H", f o r the a x i a l length test was that i t decreased with age, while The  f o r the lens, levels  ( i . e . a negative slope);  i t thickened with age ( i . e . a positive slope).  of significance <P) at which H" would be  rejected  found i n a t(17-2) test. The r e s u l t s are shown i n the Table 6. 72  were  a. 30  MALE INTRA-OCULAR  LENGTHS  -i  a  25  a  aa  3  n  a  E E  c  X  I—  2 0  A ^  '5  o  2 Legend A VITREOUS X LENS  x •  20  30  40  50  60  •  AQUEOUS  3  AXIAL  70  A G E in y e a r s  b,  FEMALE INTRA-OCULAR  LENGTHS  30 -,  a  25  £  .c t—  A A  is-  o  ~z. UJ ~"  S3  2 0  t  ir  3  A  A  A  Legend  10-  A VITREOUS X LENS  CP  X  20  a AQUEOUS  30  40  SO  3 AXIAL 60  A G E in y e a r s  Figure 20.  Intra-ocular Lengths.  73  TEST  MALE  FEMALE  Lens vs Age  < 1% 3.79 */- 0.35  < IX 3.72 •/- 0.55  < 5Y.  < 2.57. 23.16 + /- 1.05  Axial vs Age  24.31 */- 1.65  Table 6.  Significance Average  levels (P) f o r intra-ocular lengths tests. S.D. (mm) calculated from Table 5.  MALE  L = 26.79 - 0.07»A r = 0.45  L = 24.49 - 0.04»A r = 0. 52  T = 3.11 • 0.02»A r = 0.59  T = 2.76 + 0.03»A r = 0.72  Table 7.  Hence, age. case.  The This  FEMALE  Results of linear, least- squares f i t of axial length (L), and lens thickness (T) i n mm to the subject's age (A i n y r s ) . r = linear correlation c o e f f i c i e n t .  i t would seem that the c r y s t a l l i n e lens thickened  spread of the data was s i g n i f i c a n t when comparing case i s observed i n the low value of the l i n e a r  c o e f f i c i e n t s , r of the f i t s i n Table 7. 74  with. by  correlation  5.  4  R l a s m a . The  -V  e> — £ d_ "t s  four (2+1)-parameter f i t t i n g polynomials i n BLOOD.BAS  be reduced to two. most  CULT  The f i t s to the 35+2 cases i n Table 2 showed that  plasma p r o f i l e s were (as expected) best described by the  rithmic for  of the two-compartment model. t=S0 minutes p . i .  F i t #3, negative  behaviour  It produced a minimum between t=30 and  i n almost a l l cases. The one that was accepted had  a minimum that was situated i n the neighbourhood  were  loga-  or the exponential forms. The simple parabola was a poor f i t  what was expected to be a fast exponential-decay-type  time.  can  of t=60 minutes p . i .  the second-order polynomial i n 1/t,  values to a maximum i n 0. 5 < t < t " ' , 1  usually rose from the f i r s t  sampling  This resulted i n negative areas upon integration. Such r e s u l t s rejected even though the reduced chi-square might have been the  smallest among the f i t s .  FIT FUNCTION SUBJECT TYPE #0  TOTAL .  #1  .  #2  .  #3  Male Diabetic Female Diabetic  0 0  6 3  4 1  2 1  12 5  Male Normal  0  3  1  0  4  Male MS Female MS  0 1  3 7  0 3  0 2  3 13  TOTAL  1  22  9  5  37  Table 8.  Plasma f i t s .  75  Testing interaction  the at  above table i n a two-way  difference  function #1 was  between  at  t=0  The  interval,  four samples, and  t  may  ( l >  effect  was  not  The  The  the  then  the better defined the f i t .  The  replications  the  more samples taken within the  one-  However, with only the area between  treatments were  t=0  Hence,  the  taken on  the  the  intervals  [51,52] were the f i t t i n g functions.  considered. t ' ' did not have a s i g n i f i c a n t e f f e c t ; the types of 1  f i t were s i g n i f i c a n t (as shown before).  last  minute  area depends on the f i t which i s defined by  investigated.  . The  At P=10X,  interaction  functions:  have s i g n i f i c a n t effect on the integration.  , l )  was  in the blood r i s e s from  (or less in 4 diabetic subjects),  f i t was  containing t  be  best-fit  of the time at which the f i r s t blood sample was  resulting  Sex  of  the amount of dye  of samples collected.  hour p . i .  type  the best f i t . There  (injection) to a maximum i n less than one  begins to f a l l . number  the  not  s i g n i f i c a n t l y the most probable r e s u l t .  As mentioned before, zero  without  P=57. shows that the types of disease (rows) did  have s i g n i f i c a n t effect on which function was significant  classification  There did not seem to be  between the types of disease and  any  t' '. 1  above results imply that the c u r v e - f i t t i n g i n BLOOD.BAS may  shortened to save running and printing time: function(s) may  the f i r s t  be omitted from consideration.  Table 9  d i s t r i b u t i o n of the 35+2 f i t s i n the t' ' i n t e r v a l s . 1  these i n t e r v a l s were a r b i t r a r i l y chosen.  76  and  Note  the shows that  INTERVAL of t< » '  FIT FUNCTION TOTAL  in minutes p . i .  <6  6 to < 10  > 10  #0  #1  #3  0 0 0  2M 3D 2N  0 3D IN  0 2D 0  13  0 0 0  8M 3D IN  3M 2D 0  2M 0 0  19  in* 0 0  0 2D 0  0 2D 0  0 0 0  TOTAL  Table 9.  #2  21  11  37  Effect of the f i r s t blood sampling time, t M = MS; D = Diabetic; N = Normal  l1 1  LLoD  TYPE  AVERAGE  S.D.  NUMBER  Male Diabetic Female Diabetic  4.32 4. 57  1.25 1. 84  10+2 5  Male Normal  3.99  0.68  4  4.62 4. 51  3.19 2.10  3 13  4.41  1.72  35+2  Male MS Female MS FINAL  Table 10.  Average LLoDs and S.D.s. (ng. ml"' )  77  The LLoD was calculated from the background scans (0.AVG f i l e s ) in  the i n t e r v a l between 8 and 10 mm from the  the  retina.  Specifically,  i n vivo LLoD i s defined i n t h i s study as the average  tion  in  concentra-  the i n t e r v a l plus twice the root-mean-square value  of  the  types  of  This  was  Hence,  the  standard deviations of the data c o l l e c t e d i n that i n t e r v a l . An disease  analysis did  expected average  of  variance at P=257. showed that  not affect the averages i n  each  the  treatment.  because the p r e - i n j e c t i o n scans were dye-free. LLoD'.  This  was  equivalent to a concentration  of  4.41  value compares favourably with the i n v i t r o  +/-  1.72  LLoD  of  approximately 5" 1 that was estimated during concentration c a l i brations, and as stated i n [14].  5-S  A u t o f l u o r e s c e n c e  P l o t s of lens autofluorescence vs age of subject, of  disease are shown on the next page.  fits  were  slopes case;  of  found  Least-squares  f o r each disease category.  the f i t s were performed on H * :  and duration straight-line  t(n-2) t e s t s  the slope was 0  on in  the each  and on H * : the slope was p o s i t i v e i n each case. Table 11 shows  the r e s u l t s of the t e s t s against age only. These  r e s u l t s imply that lens autofluorescence increased  age [19,35]. slopes and  of  Also, the averages, i n t e r c e p t s and approximately equal the f i t s suggested that lens autofluorescence was  higher  occurred e a r l i e r i n the d i a b e t i c subjects than the normal and MS  subjects. and  with  However, t h i s trend was not c l e a r l y defined between the MS  the normal subjects as the slopes were d i f f e r e n t . 78  a.  LENS AUTOFLUORESCENCE vs AGE  300 A  250  A  C  y  200  o  in £ 150  o  z>  —1  O n <  100 -  Legend  on £ 50'  x 2<  A  10  c  A DIABETIC  ^  X NORMAL  •  20  • MS '  30 40 50 AGE in years  SO  70  b. LENS AUTOFLUORESCENCE vs DURATION 300A  A  *  250o> 200c  .c O  A  X  !<  150-  CJ  100-  X  ~r O  Ax  A  a  50-  *  A A  X A 0-  X A  X  0  5  A  &  .  ' A 0IA8ETIC X MS  10  15  DURATION in years  Figure 21.  Legend  X  Autofluorescence. 79  20  DIABETIC Sample size, n Correlation, r Average age (yrs) Average reading Standard Deviation Slope ("'.yr" ) Intercept ( ) 1  -1  P(reject  Table 11.  The  15 0.65 30.2 121.8 83. a 4.11 -2.32  H*)  NORMAL  MS  4 0. 93 32.0 55.4 20. 4 4.20 -78. 0  16 0.84 37.9 69.4 43.4 2. 93 -41. 7  17.  57.  Lens autofluorescence, S.D.  17.  (" ) vs age. 1  duration of diabetes did not have a s i g n i f i c a n t effect  lens autof luorescence (P>25V.),  on  and was not included. Autof luorescence  versus state of DRP was tested.  Number Average S.D.  Table 12.  D<1>  D(2)  DO)  8 69.94 35.36  3 184.9 73.7  4 178.3 104.0  Autofluorescence, S.D.  ("') vs DRP states.  t - t e s t s on differences showed that D(l) and D(2), D<1) and D(3) were  s i g n i f i c a n t l y d i f f e r e n t (P=107.);  while D<2) and D(3) were  not  (P>257.). Note that the S.D.s were sizeable f r a c t i o n s of the averages, i.e.,  there was a large variation from case to case. (An analysis of  variance at P=2.5% showed that at least two of the means were not the same.) One  conclusion  i s that  60  lens  autofluorescence  increased  significantly These  results  r=0.65, the  vith  progression from no DRP to severe  partly  explain  the  lov  DRP  correlation  [19,35],  coefficient,  f o r the c u r v e - f i t t i n g i n Table 11. D i a b e t i c subjects betveen  age of 18 and 27 years v i t h severe DRP reduced the  goodness  of  the s t r a i g h t - l i n e f i t . Hovever, i t has not been e x p l i c i t l y shovn that the increase i n lens autofluorescence i s due to the thickening of the lens  vith  age.  This requires a much l a r g e r sample s i z e  of  normal  subjects spanning a large age range.  MS STATES  AVERAGE  S.D.  NUMBER  MS(1) MS(2)  69.96 68. 85  48. 93 40.45  8 8  MS (3) MS<4)  47. 38 97.71  27.69 44. 93  9 7  MS<5) MS(6) MS(7)  69. 74 73.03 67. 30  54.35 49.66 29- 60  7 3 3  Table 13.  Autofluorescence, S.D. (" 1 ) vs MS s t a t e s .  There vas no s i g n i f i c a n t difference betveen the means of and MS(2),  or that of MS<5),  MS(6) and MS<7),  MS(1)  (P>25*4). Hovever, the  means of MS(3) and MS(4) vere s i g n i f i c a n t l y d i f f e r e n t (P=2.57.).  Note  the large S.D. Although the trend of increasing autofluorescence v i t h age seen recent  i n Table 11, clinical  v i t h diabetes.  vas.  the c l a s s i f i c a t i o n i n terms of graded s e v e r i t y of a c t i v i t y did not shov trends s i m i l a r to  the  cases  The d i s t r i b u t i o n of those older subjects among the MS 81  groups had "biased" the r e s u l t s .  No trends were observed except that  autofluorescence was higher than f o r normals. The  trends  researchers' values this  established  findings.  thus  Hovever,  f a r are i n  comparing  here are much lover than from [351. is  average  the  angle  vith  other  readings,  the  One possible reason  for  that the instrument c h a r a c t e r i s t i c s and example,  accord  betveen  calibrations  beam  and  vere  different.  For  probe  vas  different.  Hovever, i t i s more l i k e l y that alignment errors vere the  cause of the differences.  5 .  "7  P r o f i l e s The  (Figure are  effects  of  CR t a i l i n g s vere studied i n a  D(l) subject.  22.) Some of the scanning problems discussed i n Section  demonstrated  2.1  i n these p r o f i l e s .  The excitation beam i n the 2-minute scan vas probably p a r t i a l l y clipped  by  the i r i s as suggested by the plateau  f i r s t 2 mm from the retina.  The t a i l i n g s hovever,  shovn  vithin  the  coincide with the  4-minute scan f o r distances greater than 2 mm from the retina.  that  The  4-minute  was  located  scan peaked at about the position of the visually.  Note that the t a i l i n g s  at  retina  this  time  persisted well into the vitreous where, f o r t h i s subject at t h i s p . i . time, no dye was expected to have penetrated. The the  6-minute p r o f i l e shows the d i f f i c u l t y i n v i s u a l l y locating  retina.  The peak i s approximately 1 mm anterior to the  retina.  This p a r t i c u l a r bolus alignment error can be corrected by s e t t i n g the zero  at  the CR peaks,  but f o r l a t e r measurement scans  82  the  actual  100CH  DISTANCE FROM RETINA in m m  Figure 22.  position  of  Bolus effects.  the retina by CR peaks can not be c l e a r l y  defined  in  practice or i n theory. Figures  23a-23g show the evolution of the dye p r o f i l e  posterior vitreous of subjects i n various c l a s s i f i c a t i o n s . the  v e r t i c a l log scales are different.  The important  i n the  Note that  points are as  follows a) the prominance of the bolus p r o f i l e s and CR peaks. b) the change of slope with time about 3mm. c) alignment and peak s h i f t s anteriorly. d) the difference i n the concentrations f a r t h e r from the retina. The  profiles  of  the normal (Figure 23a) are  83  "noisier"  than  Legend A 4-UINUTE X 37-UINUTE a S2-UINUTE Legend A  cn c _c  X  s-uiNurc  g  Q  0-UINUIE  a  SO-MINUTE  <  A X  ^  x^  ( V  z  o o  O <J  —I  1 : 1 rt 2 J 4 5 DISTANCE FROM RETINA in mm  0  Legend A 3-UINUTE X 31-UINUTE  x g^-^o^  q_p  aar  1 2 3 * 3 OISTANCE "ROM RETINA in mm  S  b. DU) - no DRP.  a. Normal.  &<.  a  SI-UINIJTE  a  O-UINUTE  A X • a  1 0 0  c  1  Legend 4-UINU1E 51-UINUTE 53-UINUTE 0-MINUTC  -S  < 10'  o 5?  p:  CJ  z o  <J  ,  *x  a a a a  Z UJ CJ  c  a  c c  x 32-uiNlirE  z  X  ****  .,  o  _  31.,  "HOT, 4  0  1 2 3 4 5 DISTANCE FROM RETINA in mm  d. D(3) - severe DRP.  c. D<2) - moderate DRP.  Figure 23.  1 2 3 4 5 DISTANCE FROM RETINA in mm  Sample p r o f i l e s .  84  ^  Legend Legend  x 31-uiNUTE  A 3-uiNUTE X 30-UINUTE • SO-UINUTE 9 O-UINUTE  =  < es  <  CE  Hi U1 (TA A  U  z o o  0 SI-MINUTE  o> c  a  O-UINUTE  " * ^ x  to-  o z o u 3  1  3  2  *  5  1  DISTANCE FROM RETINA in mm  2  e. Stable MS.  Legend X 27-UINUTE  O 60-MINUTE 8 0-MINUTE  'oo J  o  i=  10  < O  0  1  2  3  4  5  DISTANCE FROM RETINA in m m  f. Liquefied vitreous.  Figure 23.  *  f. Relapsing MS.  A 5-MINUTE  5  3  S  DISTANCE "ROW RETINA in mm  Sample p r o f i l e s (continued),  85  others as one might expect from the lower concentrations of dye. A l l profiles  more or less coincide at 6mm unlike f o r others,  the DO)  especially  and the MS cases i n Figures 23f,g.  The vitreous  special case of the female,  MS subject with the l i q u e f i e d  i n Figure 23g i s worth noting.  The gradient of the  1-hour  p r o f i l e (at the posterior vitreous) i s small compared to the bolus at various positions from the retina. well  defined.  •flat",  However,  as  the  The 60-minute CR peak i s thus not p r o f i l e about the 3-mm  point  was  misalignment should not produce large errors i n t h i s special  case.  However,  shows  that  regardless of the state of the vitreous,  a large amount of dye had indeed entered  Figure 23g  the posterior  vitreous through the BRB. Figure  24a  compares the 1-hour scans of two stable,  MS  sub-  j e c t s ' p r o f i l e s . The male, MS(4) p r o f i l e i s very s i m i l a r i n slope and magnitude to the MSO),  female subject.  normal subject's p r o f i l e .  Both are elevated above the  The plateau between 2 and 3 mm i s probably  due to movement by the subject during the scan C321. The male subject had no p e r i p h l e b i t i s ; the female subject was not examined f o r this. Figure subjects, latter  24b  compares  the  1-hour  scans  of  with active and inactive p e r i p h l e b i t i s  subject  was  the person with the  two  female,  respectively.  liquefied  vitreous.  MS The Both  belonged to the MS(2) and the MS(4) groups. Note that the MS p r o f i l e s are c l e a r l y above the normal. Figures 25a,b,c compare a male 1-hour p r o f i l e s i n the D(l), magnitude  of  and a female diabetic person's  D(2) and DO)  groups  respectively. The  leakage ( v e r t i c a l axis) i s progressively greater 86  from  cn  X*  c  A  z o  *  ^  A  CD  • *S "A  z o<  A, A  cc  A  NORMAL  X  145.MS  •  U'.MS  AA A  A" *  . ^ A  A  *X  LLJ  AA  Legend  x*x*x  z o  I—  z UJ o z o  ^bg«S  O  A  2  3  4  A"A  Legend  o A<£  1  CP  5  A  NORMAL  X  246.MS  Q  247.MS  A  A  V  A  A-  AAA  A  A ^  1  DISTANCE FROM RETINA in mm  AA  2  3  4  5  DISTANCE FROM RETINA in mm  Relapsing-Remitting vs Relapsing-Progressive.  b.  Relapsing-Progressive vs l i q u e f i e d vitreous.  Figure 24. Comparison of MS p r o f i l e s . (Refer to Table 14 f o r number codes.)  D<1)  to DO).  This demonstrates good correspondence  severity  of  leakage  i n Figure 25a of D(l) p r o f i l e s may have been accentuated  control  than  5  The differences i n the  clinical  grading  poor  of DRP.  vith  magnitude  of diabetes i n one case despite the presence  microaneurysms.  The leakage i s s i m i l a r to that of a  of  of by less  normal  subject vhen there i s good control of diabetes. Figure 26 compares the 1-hour scans of the following subjects: a)  a male subject i n the MS<1), MS(4) and MS(5) groups.  b)  a female subject i n the MS(2), MS(4) and MS(6) groups.  c)  a female subject i n the D(l) group.  d)  a female subject i n the D(2) group.  e)  a male subject i n the DO)  These  illustrations  group.  of p r o f i l e s are intended  87  to  demonstrate  Legend A NORMAL  CO  X 1F.DIAB  c c  z o  *x  A  a:  A  o z o  a. Male vs Female D(l),  ^ ,A  XA^^AA A A* AAA  1  2  4  3  5  DISTANCE FROM RETINA in mm'  Legend A  NORMAL  X 2F.DIAS •  z o  xxx  <  2M.0IAB  b. Male vs Female 0(2).  0  \  O  A~" ^  A  A  AA A A I S  A  AAA 1  2  4  3  5  DISTANCE FROM RETINA in mm  Legend  z o  K  t— Z UJ  A ^ A A A  %3? • a  V  o z o  ^  c. Male vs Female D(3).  AA AAA  1  2  3  4  5  DISTANCE FROM RETINA in mm  Figure 25.  Comparison of diabetic subjects' p r o f i l e s .  88  Legend  4  g>  100  < z  UJ  a.  u  o  1  0  I  2  3  4.  5  JJCA  6  DISTANCE FROM RETINA in mm  Figure 26. Overall comparison. (Refer to Table 14 f o r number codes.)  different  magnitudes  of leakage v i t h i n MS  and  diabetes.  Profiles  provide q u a l i t a t i v e comparisons of the i n t e g r i t y of the BRB - such as v i s i b l y d i f f e r e n t gradients (at some point),  or, that one l i e s above  or belov the other. In the calculations of PR3 or P , 1  the  r e s u l t s of the plasma integral may produce quite d i f f e r e n t quan-  t i t a t i v e descriptions of the  5 .  S  D i f f u s i o n  BRB.  Cons-tern-t  The d i f f u s i o n constant, 14  the d i v i s i o n by  gives  D vas calculated by SLOPES.BAS.  Table  the results of averaging over a l l measurement scans  made.  Note that D and a l l calculations pertaining to i t are alvays given i n units of *10-' em's" . 1  Least-squares f i t s to D = A + B»(age), and D = X • vere carried out on each disease group v i t h H* slope < 0 or > 0,  i.e.,  Y»(duration).  : slope = 0, and, H  A  :  one-sided t - t e s t s depending on the c o e f f i -  cient, B or Y of the f i t s . 89  AGE  DIABETIC  18 19 20  4.44 (Fl) 0. 85 (F3)  21 22  3.73 3.69 9. 78 0. 55  NORMAL  MS  4. 42 (F235)  23 25 26 27 28  (Ml) (Ml) (Ml) (Ml)  1. 35 (F23?) 5.18 (F135)  2. 94 (F3) 7.38 (M) 7.18 2. 54 5.22 1.84  29 30 31 32 35  (Ml) (M3) (Ml) (F2)  6. 42 (F13?)  6.02 (Ml)  6.72 (M) 4.12 (M)  3.58 (M2)  37 38 39  15.15 (M>  44 49 56 57 59  17.13 (F135)  3.62 7.11 9. 33 5.22 4. 34 11. 49  (M137) (M145) (F236) (F247) (F246) (M135)  29.90» (F247) 8.72 (F247) 27. 59 (F135) 2.69 (F2) 18.68 (F145) 13.13 (F24?)  64  2. 31 (M3)  AVERAGE S.D.  3.83 2. 44  NUMBER  10M:5F  8.34 4. 75 4M:0F  10.86 8. 49 3M:13F  D averaged over a l l measurement scans. CODES : (XA) f o r diabetics; (XBCD) f o r MS where X = Male or Female A = 1,2,3 f o r D(i),D(2), D(3) respectively; B = 1,2 f o r MS(1),MS(2) respectively; C = 3,4 f o r MS(3),MS(4) respectively; and, D = 5,6,7 f o r MS(5),MS(6), MS(7) respectively. D = ? means subject not examined. » i s the subject with l i q u e f i e d vitreous.  90  DIABETIC  NORMAL  Sample Size Coefficient, A Coefficient, B Correlation, r  15 4.82 -0.03 0.18  4 -16.06 0. 76 0. 73  16 -4.96 0.42 0.62  P(reject H«)  > 25%  257.  17.  Sample Size Coefficient, X Coefficient, Y Correlation, r  13 5.84 -0.22 0. 53  P(reject H«)  35 -1.08 0.25 0.46 < 0.57.  57.  Tests of D vs age and duration.  AVERAGE + /- S.D. D<1) D(2) D(3) Male Diabetic Female Diabetic  TOTAL  9 4.61 1.14 0.63  57.  Table 15.  MS  NUMBER  P  5. 08 2.70 2.16 4.46 2.55  2. 73 0.87 0. 91 2.67 1.33  8 3 4 10 5  257. 2. 57. 0. 57. 107. 0.57.  3.83  2. 44  15  0. 57.  8.34  4.75  4  257.  12. 15 9.58 9.61 12. 47 13.08 6.84 11.87 7.41 11. 66  8. 33 9. 01 8.25 9.17 8. 50 2.50 12.21 3. 94 9.15  8 8 9 7 7 2 4 3 13  57. 257. 257. 107. 57. > 257. 257. > 257. 2. 57.  MS AVERAGE  19.86  8. 49  16  2.57.  FINAL AVERAGE  7. 56  6.90  35  97.  DIABETIC AVERAGE Male Normal MSU) MS(2) MS<3) MS(4) MS<5) MS<6) MS<7) Male MS Female MS  Table 16.  Diffusion constant, D by sex-disease states. 91  The with  r e s u l t s i n Tables 14 and 15 show that D tended to increase  age,  This  numerically betic  result was dominated by the larger MS  more older subjects.  subjects  Also,  sample  with  the B-coefficient f o r dia-  was negative but not s i g n i f i c a n t l y  so.  Others  had  also noted t h i s trend but were unable to prove i t s t a t i s t i c a l l y C533. The tests of D against the duration of diabetes (Table 15) were inconclusive  because the sample s i z e s were small,  c i e n t s f o r diabetes and MS were of opposite signs,  and the Y - c o e f f i showing  opposite  trends.  This may be due to intra-group v a r i a t i o n i n severity. Larger  samples  of each disease state are needed to establish the  of  trend.  any  Note the s i g n i f i c a n t scatter of data as  existence  the  corre-  l a t i o n c o e f f i c i e n t s , r, are not close to 1. (Figure 27.) The  extreme right column i n Table 16 shows the r e s u l t s of the  tests f o r H* where  D"=6  Despite result  :  average D i n each group = D" ,  [20] i s the d i f f u s i o n constant of the  the  large spread of the data i n each  dye  group,  i n water.  one  notable  was that the D's f o r diabetics were s i g n i f i c a n t l y lower  than  D", f o r t h i s sample. The reason f o r t h i s i s not known. Similarly, the explanation D(l)  f o r the D(2) and the D(3) averages being half  i s not apparent.  that  I t should be noted that the D(l) sample  of was  twice the s i z e of D(2) or D(3). In  contrast,  higher than D . M  the older, fied  the average D f o r MS subjects was  As stated before,  MS subjects,  significantly,  the high values came from most of  notably, the female, MS subject with lique-  vitreous who had the highest value.  92  This r e s u l t may be due  to  Lscsnd 4 M-auacnc •  <&  Legend  23.0-  A NORMAL  l/l £ u 20.0-  NORMAL  X  M-MS  C  F-MS  Z  <  z  o a 5 •£  C u z c  13.0-  ta.o-  3.0-  DIF  u.  :0  20  JO AGZ  *a ;n  SO  50  3.0-  0.0-  70  20  20  years  a. Diabetic subjects. Figure 27.  40 AGZ  in  30  SO  yecrs  b. MS subjects. Diffusion ocnstant vs age.  the effect of mechanical mixing on the difference equation, Eq. 32. Other tests on the results i n Table IS shoved that the value of D  for  D(l) vas s i g n i f i c a n t l y d i f f e r e n t from those of D(2) and  (P<5/£).  There  vas no s i g n i f i c a n t difference betveen D(2)  and  D(3) D(3)  <P>257.). Betveen MSU)  and MS(2),  MS(3)  and MS(4), and MSC5), MS(6)  and  MS(7), there vas no difference (P>25/C). These r e s u l t s vere due to the large  spread  of  data i n each group v i t h  high  values  distributed  throughout a l l groups. The others  diffusion  coefficient  i n the normal  eye  vas  to be  a)  13.3  >/- 6.8 and 11.9 +/- 5.4  b)  13.2  +/- 4.3  (Ogura, et a l . [531),  c)  7.4  •/- 3.4  (Lund-Andersen, et a l . [54]).  93  (Chahal, et a l . [23]),  found  by  E ° c  20.0 Legend  z <  «. DIABETIC  (/I  x MS  z o o  1/1  3  0  5  10  15  20  DURATION OF DISEASE in years  Figure 28. For  diabetic eyes,  Diffusion constant vs duration.  D = 9.6 +/- 2.0 [54].  Note that the large S.D.s  allow f o r much overlap. Comparing these results to those i n Tables 15 and 16, f o r P=55C, there  was  derived  no difference between the average found  by others f o r normals.  here  and  those  Individual r e s u l t s i n Table 15  were  within the range i n (c); but, somewhat lower than those values i n (a) and (b). Comparing the averages of any diabetic group, or, of individual cases,  a l l were  found to be s i g n i f i c a n t l y lower (P<0.5%) than  the  value stated by Lund-Andersen, et a l [54], Many  of  the D values f o r MS subjects were within  values f o r normal and diabetic persons. fied  the  The exception was the lique-  vitreous case f o r which D was higher than a l l others.  were also elevated f o r several older MS subjects.  D values  No explanation i n  terms of c l i n i c a l a c t i v i t y i s known. (Refer to Table 14.) 94  quoted  The r e s u l t s of D i n Table 14 were derived f o r p r o f i l e s by  RET  only.  values.  As  Misalignment  the  might account f o r some of  the  average over a l l measurement scans was  should a l l e v i a t e the alignment errors.  aligned extreme  used,  this  Also, no correlation was seen  between D and the PR3 values studied below.  5 -  9  P e n e t r a t i o n Recall  that PR3  R a t i o  (in units of »10"* s" ) was calculated i n 1  methods of alignment by the programme, CRP. In  (Refer some  changed same.  to Section 3.4.)  cases,  PR3  and,  by  The r e s u l t s are presented i n Table 17.  was more than halved when  alignment  was  i t remained approximately  the  changes are evidence of the d i f f i c u l t y of locating  the  from RET to CRP; These  C/VAZ.BAS: . by RET,  two  i n others,  the  r e t i n a by sight. A case i n point was the 22-year old,  MS subject whose  1-hour,  PR3 value was 48.2 by RET - the highest of a l l MS subjects; while, by CRP,  i t f e l l to 13.1.  that  the CR peak was more than 2 mm from the located zero  Although phlebitis,  Her fluorescein p r o f i l e at 1-hour p . i . showed position.  she was not examined f o r the presence or a c t i v i t y of her  peri-  p r o f i l e s were not d i s t i n c t l y d i f f e r e n t from other MS  subjects i n the MS(2)  and MS(3) groups,  but,  were when compared  to  the MS(6) category. Another noteworthy result i s that of the 32-year old normal who • i s the brother of the 35-year old, about 3 times that of other normals. scanning,  was i n relapse,  MS subject.  His PR3 values  His s i s t e r , who,  at the time of  had lower PR3 values than h i s 95  were  (in either  DIABETIC nut  RET  18 19 20 21 22 23 25 26 27 28 29 30 31 32 35  CRP  64  RET  CRP  RET  CRP  2.4 (Fl) 4.8 294.2 (F3) 103.1 16.2 (F235) 15.7 (Mi) (Ml) (Ml) (Ml)  18. 0 12.1 32.7 16.6  144.4 (F3)  63.8  29.5 11.2 10. 0 1.4  48.2 (F23?) 13.1  30. 6 (F135) 17.6 6.2 (M) 1458.5 27.5 10.4 11.1  3.1  (M3) 886.8 (Mi) 22.1 (Ml) 6.5 (F2) 8.3  8.5 (Ml)  14.6  30. 0 (M2)  24. 6  37 38 39 44 49 56 57 59  MS  NORMAL  15.4 (M) 15.8 32.3 (M) 22.3  8.9  4.0 11.2 25.2 12. 2 37.0 14. 7  (M137) 1.6 (M145) 7.7 (F236) 18.2 (F247) 16. 2 (F246) 33. 5 (M135) 6. 7  31. 9» (F247) •29.0 9. 5 (F247) 9. 5 9.9 (F135) 10.1  12.9  2.4 (F145) 5.2 (F24?) 138.2 (M3)  No.  4.2  22.8 (F13?) 11.7  8.4 (M)  13. 0 (F2)  4.6 (F135)  1.9 5.8  66.1  10M:5F  4M:0F  Table 17.  3M:13F  PR3 averaged over a l l 55-70 minute scans, after background subtraction only. CODES : (XA) f o r diabetics; (XBCD) f o r MS vhere X = Male or Female A = 1,2,3 f o r D(1),D(2),D<3) respectively; B = 1,2 f o r MS(1),MS(2) respectively; C = 3,4 f o r MS(3),MS<4) respectively; and, D = 5,6,7 f o r MS<5),MS(6),MS(7) respectively. D = ? means subject not examined. • i s the subject with l i q u e f i e d vitreous.  96  Legend  Legend a  NORMAL  x  l-o  4  2-0 a 3-0  .0 2-0 a j-o 3  1-0  x  5  NORMAL  a  MS  O  MS  5  z o x ""a  z o  3  z  z  UJ  a.  UJ 0.  30 »0 50 AGE in years  60  30 40 so AGE in years  a. By RET.  b. By CRP.  Figure 29.  alignment)!  This  calibrations  so  Penetration Ratio.  anomaly might have been due to improper instrument  (settings) at  that  time;  otherwise,  i t cannot  be  explained. This subject (and h i s s i s t e r ) w i l l have to be recalled f o r further testing.  His reading vas omitted from analysis.  other data such as F-number calculations vere s t i l l The by  MS(6)  v i t h the l i q u e f i e d vitreous,  •clustering*.  vere This  The case of (F247),  the subject  also had a high PR3. A l l other MS c l a s -  distributed throughout the order v i t h implied  sepa-  The f i r s t noticeable point vas that a l l  subjects had higher values.  sifications  anticipated  PR3 results i n ascending order (by RET and by CRP  f o r MS subjects.  that  admissible.)  r e s u l t s of s t a t i s t i c a l - i n f e r e n c e t e s t i n g vere  ranking  rately)  (Note  that there vas no  detected  betveen MS(1) and MS(2), MS(3) and MS<4), MS(5) and MS(7).  97  no  obvious  difference  The male,  highest PR3 results calculated was that of a 22-year old,  D(3) subject.  Although he has severe DRP,  h i s reading  was  about 10 times higher than other D(3) subjects which may be an i n t r a group  variation.  levels  The 1-hour p r o f i l e showed  of dye i n the vitreous.  indisputable, elevated  His PR3 value was also omitted  from  a l l testing. The ordering of diabetic PR3 r e s u l t s showed that the r e s u l t s of the  D<3)  group  obviously  were consistently the highest  values.  Cases  with  high amounts of leakage were c l e a r l y detected (and detec-  table) by t h i s fluorophotometer. This " j u s t i f i e s " the omission.of.the above D(3) case because the other D(3) results, taken i n d i v i d u a l l y or together,  were already s i g n i f i c a n t l y higher than those of the other  two diabetic groups.  1 1  •'  RET  CRP  GROUP MEAN +/- S.D.  #  MEAN +/- S.D.  #  D(l) D(2) D(3)  12.6 18.0 192.3  10.5 10.4 88.3  8 3 3  15.9 15.3 77.7  8.9 8.4 22.1  8 3 3  NORMAL  10.0  4.8  3  9.3  6.4  3  MSQ) MS(2)  12.5 19.6  9.9 12.0  8 7  7.7 17.6  5.4 9.4  8 8  MS(3) MS(4)  16.0 15.6  9.7 13.4  8 7  11.0 14.8  5.9 12.1  9 7  MS(5) MS<6) MS(7)  12.8 31.1 14.4  9.3 8.3 12.2  7 2 4  9.1 25.8 14.1  5.8 10.8 11.6  7 2 4  Table 18.  PR3 Average •/- S.D. of the various groups.  98  The DU) and D(2) c l a s s i f i c a t i o n s , MS(3), RET  MS(4) groups,  l i k e the MS(1),  MS(2), and  did not d i s t i n c t l y separate out i n either the  or the CRP sorts.  This could imply that intra-group and  group  fluctuations vere s i g n i f i c a n t .  rithm  not optimized f o r such c a l c u l a t i o n s could also have  inter-  Poor alignment and/or an algoprevented  the appearance of any expected order. PR3* r e s u l t s by SLOPES.BAS ordered i n the same manner as by RET.  those  This vas expected because SLOPES.BAS vas v r i t t e n to approxi-  mate RET results by c u r v e - f i t t i n g .  (Refer to Section 3.4.) The order  f o r the D<3) group vas exactly the same but the PR3 values vere about 10'/. greater than the PR3* values. vere  neither  sign test at  On average,  hovever,  PR3» values  alvays greater than nor alvays lover than PR3 (by the P=5V.).  PR3* results from SLOPES. BAS could then be  to check C/VAZ.BAS's RET PR3 results.  used  Either set of r e s u l t s could be  consistently used to represent the penetration r a t i o vhen only alignment  by  RET vas considered;  computing time could be  shortened  by  choosing to run one of the tvo programmes only. It i s observed from Tables 17 and 18,  that most individual and  average results of diabetic and MS subjects vere higher than those of the  normals,  betveen  despite the large S.D.s.  Although age-matching  tests  the members of each group vere not possible because of the  small sample sizes, such trends demonstrated that differences betveen groups and indivduals existed (for t h i s sample) and vere detected! Tests  of  the s i g n i f i c a n c e of the differences betveen the PR3  means betveen any tvo groups vere carried out. vas used to test 99  Analysis of  variance  H*  : the means of any 2 groups were the same,  i . e . t h e i r difference vas 0, against H  a  : the means of any 2 groups vere different,  i . e . t h e i r difference vas not 0. The entries  P(reject H*) are shovn i n the respective tables belov. The i n the upper triangles are f o r RET alignment.  Those i n the  lover t r i a n g l e s are for CRP alignment.  BY RET  B Y C R P  NORM  DU)  DO)  NORM  »•*»  > 25  > 25  2.5  D(l)  > 25  » »* *  > 25  < .5  D<2)  > 25  > 25  **»#  5  DO)  1  < .5  2.5  Table 19.  only the DO) group vas c l e a r l y and s i g n i -  f i c a n t l y d i f f e r e n t from a l l other groups. D(2) ranged  respectively. occurred  from 1.4 to 30,  These  • *•»  Significance l e v e l (X) to reject H* betveen diabetic and normal groups.  In the above table,  and  DO)  But, the r e s u l t s in  DU)  and 6.5 to 32.7 f o r RET and CRP  indicate that breakdovn of the BRB had already  and vas detected in subjects vithout signs of DRP but  vere.  at the early stages [4,5]. The  only  s i g n i f i c a n t difference f o r the groups i n  belov i s that betveen MSU) and MS(2)  100  i n CRP.  Table 20a  In Table 20b,  MS(6) i s  significantly  different  Betveen other groups,  from MS<5) and normal but not from  there i s no s i g n i f i c a n t difference.  MS(7).  Table  18  shovs the large S.D.s of these groups.  BY RET  NORMAL B Y  NORMAL  MS(1)  **»*»»  > 25 ••*»•  MS<1)  C R P  > 25  MS(3)  MS(4)  25  > 25  > 25  25  *****  *****  MS(2)  2.5  *****  *****  • *•*.*  > 25  *****  *****  *****  > 25  > 25  *****  • •••*  > 25  *****  MS(2)  25  MS (3) MS(4)  BY RET (b) NORMAL B Y C R P  NORMAL  reason  MS(6)  MS<7)  > 25  5  > 25  • ••••  5  > 25  MS(5)  > 25  MS(6)  25  2.5  MS(7)  > 25  > 25  Table 20.  One  MS (5)  • ••*» > 25  25 • ••••  Significance l e v e l (.'/.) to r e j e c t H* betveen MS and normal groups.  f o r the great dispersion of data i n MS<7)  is  the  placement of the case of the l i q u e f i e d vitreous. Her c l a s s i f i c a t i o n s , place her into the respective categories, results  but i t i s not clear i f her  should be included at a l l because of her unique 101  case.  When  MS(7)  i a tested without i t ,  against MS(6)  i n RET,  P f a l l s to  2.57.  from 25%. A l l other comparisons remain the same. From the averages and S.D.s between the two diseases,  i n Table 18, i t i s easy to see that  there i s no s i g n i f i c a n t difference between  D(l)-D(2), MSU)-MS<2) and MS(3)-MS<4) c l a s s i f i c a t i o n s when comparing within or between these groups. D(3) i s , of course, very much greater then  a l l other groups.  cantly  The only set which i s " i n t e r n a l l y "  d i s t i n c t i s the MS(5)-MS(6)-MS'7) set.  Hence,  signifi-  i t i s tested  against D(i) and D(2) only.  R E T C R P  Table 21.  MS(5)  MS (6)  D(l)  > 25  10  > 25  D(2)  > 25  25  > 25  D<1)  > 25  25  > 25  D(2)  25  > 25  > 25  MS<7)  Significance l e v e l (%) to r e j e c t H" between diabetic and MS states.  There i s no s i g n i f i c a n t difference (at P=5%) and there  a l l MS is  groups.  between DU),  If these treatments and r e s u l t s  are  correct,  no s i g n i f i c a n t difference between an MS subject's PR3  that of a diabetic with n i l to moderate DRP,  or a  normal.  D(2)  and  However,  i n d i v i d u a l variations i n PR3 values (Table 17) should be noted. The applied,  results with  are  similar when the CR  bolus  corrections are  and without the correcting peak-to-peak 102  ratio.  (See  Section  2.3.)  It  i s hence not possible to study  the  effects  of  applying these corrections. The r e s u l t s thus f a r indicate the problems that are inherent i n the VF technique. in  the  Assuming no instrumental errors,  positioning of the retina and the alignment of  analysis reduce the certainty of the PR3 The MS(3)  above  and MS(4)  profiles  in  MS(2),  and  results.  results and tests show that MS(1) cannot be d i f f e r e n t i a t e d .  variations within each group. is  the d i f f i c u l t i e s  and  This i s due to the  It does not mean that the VF  not applicable to MS as the subjects i n MS(6),  technique  with active p e r i -  p h l e b i t i s , were discernible from others of the MS<5> and MS(7) as  was  the  case for diabetes where the severity  corresponded to the severity of  large  of  groups  leakage  also  DRP.  Comparing the results for normals i n Table 17 to other i n v e s t i gators' results,  which are from 3.5 to 5.3  [32], i t i s seen that the  PR3 values calculated here are within or above t h i s range. risons are available f o r MS PR3. in  some  However,  No compa-  abnormal leakage was seen  MS subjects other than the two with  active  periphlebitis.  (Refer to Table 17.) These elevated PR3 values cannot be explained by retinal tion).  vasculature They  imply  appearance (photographs and that the VF system may  technique to detect s u b c l i n i c a l a c t i v i t y . only the  clinical  be useful as a  sensitive  However, the present  included a small sample of subjects and i s not able clinical  examina-  to  study relate  gradings of a c t i v i t y of MS or the current a c t i v i t y  the time of the procedure.  103  at  5 . 1 0  L U N D . B A S  R e s u l t s  LUND.BAS was not tested on a l l subject. expansion  F i r s t l y , the gradient-  algorithm i s slow (on this computer),  the i n i t i a l i z i n g estimates of P and D.  Convergence i s slow i f any of  the  input values are f a r from the "true" values.  3.4  and [33,343.) Secondly,  when  and i s dependent on  (Refer to  Section  the conditions set to halt calculations  the reduced chi-square value begins to diverge from a  minimum,  or i s less than 1, are not amply stringent i n terms of convergence to a  final  final,  solution set. best  The programme outputs the  residues  of  the  f i t but plot outputs to v i s u a l l y check the answers  are  not (yet) available. Several  outputs were returned on t h i s subprogramme  subjects tested.  They were the 27- and 38-year old normals,  38-year old MS subject i n relapse. (See Table Table  22  •10-•  cm.s .  that  were  -1  accepted  for  and  the  17.)  shows the case-by-case results. D remains i n units of »10-*  three  The units of P  cm s-'. a  are  Only data points  between 1.5 mm from the retina and the mid-vitreous  f o r c u r v e - f i t t i n g to Eq.  18 (Section 2.4).  were  The number  of  data-points that was accepted and f i t t e d i n each case i s shown i n the extreme  r i g h t column.  The results of P values from  other  investi-  gators f o r normals and diabetic subjects are also included. In comparison with published results, the ones obtained i n t h i s study are just within the range or less than those i n the references. The d i f f u s i o n constants, D, also follow the same trends when compared to those calculated i n the previous section. of the quoted r e s u l t s should be noted. 104  Again,  the large  S.D.s  INITIAL ESTIMATES  FINAL FITS CHI-SQ.  P  D  P  #  D  27-year old NORMAL: 60.RET 10.0 7.9 6.8 7.0 20.8 2.0  2.7 1.6  4.78 3.62  15 15  27-year old NORMAL: 60.CRP 2.8 10.0 6.6  5.9  1.13  16  27-year old NORMAL: 68.RET 4.0 10. 0 6.6  5.0  1.44  14  38-year old NORMAL: 64.CRP 5.3 0.5 6.6  5.1  0.70  20  38-year old MS, F246 : 61.CV2 10.0 10.0 8.1  14.4  0.63  38-year old MS, F246 : 61.CV4 7.7 10.0 10.0  16. 0  0.72  ••  References: NORMAL P -values 0 by Lund-Andersen , et a l [54] a) 11.0 •/- 4. b)  30.0 +/- 8.3  by Ogura , et a l[53]  c)  7.2 •/- 4. 4  by Zeimer, et a l [29]  d)  19.1 +/- 9.4  by Chahal, et a l [23]  References: DIABETIC P-values 71.0 •/- 38.0 by Lund-Andersen , et a l [54]  Table 22.  If  Results of LUND.BAS. »* means unavailable.  the algorithm of LUND.BAS i s to be the adopted  which d i f f e r e n t investigators compare P and D values,  method by  i t i s apparent .  that  the computational conditions on the reduced chi-square i n LUND  . BAS  must be more r e s t r i c t i v e and s e l e c t i v e .  output plots must also be implemented.  105  Double-checking  with  5 .  1 1  O - t h e r  P,  P a r a m e t e r s  the permeability index was also calculated by  1  (Refer to Section 3.4.) The results were, often  turned  out negative and  were  probably due to the algorithm i t s e l f . close  however,  rejected.  C/VAZ.BAS.  not useful. They This  failure  was  The necessity to integrate very  to the retina or to find an approximation when integrating i n  that region was machine-(AR-)dependent [32]. Another BAB.  penetration r a t i o not metioned thus f a r i s that of the  This r e s u l t was not investigated because the source of  leakage  was the i r i s and c i l i a r y body. The diamond does not scan close to the source of leakage and the models used i n the algorithms employed here are  not  applicable.  Another point i s that misalignment errors are  greater f a r t h e r away from the retina.  (See Figure 5.) Hence, a "PR3"  cannot be calculated f o r the BAB. These conclusions were borne out by SLOPES. BAS  which includes such a c a l c u l a t i o n at 3 mm from  the pos-  t e r i o r surface of the lens. No c o r r e l a t i o n s i n any of the groups were found. Two data.  other performance parameters were calculated from  The f i r s t was the i n vivo r e p r o d u c i b i l i t y (R).  subject  The d e f i n i t i o n  in Eq. 27 i n Section 2.5 was changed as t h i s parameter was calculated from two sources. One way to test R i s to take scans within 3 minutes of  one another;  the averages about c e r t a i n regions of each  profile  are found. R i s then defined as: R  =  100 « I a(x,t») - a<x,t ) I / J a(x,t») + a<x,t") } %, s  where a(x,t') i s the average about x of the t -minute p . i . p r o f i l e . R 1  106  i s simply half the deviation from the average divided by the average. The alternative i s to use the S.D. i n the numerator, but as there are only tvo entries at each calculation, t h i s vas thought to be unnecessary. Note that small R-values imply good r e p r o d u c i b i l i t y . The region about the mid-vitreous of measurement scans taken at t>55  minutes  because later  p . i . vas selected.  The 3-mm i n t e r v a l vas not used  of the influence of t a i l i n g s or large dye concentrations at p . i . times,  vhen there i s leakage.  Further,  no mixing  vas  expected at the mid-vitreous at these times. Only CRP-aligned  p r o f i l e s vere considered oving to the problems  i n locating the retina surface.  Hovever, the f i n a l R vas the average  from a l l scans made after 55 minutes p . i . made v i t h i n 3 minutes from the f i r s t .  for vhich another scan vas  R i s a systems c h a r a c t e r i s t i c ,  and no tests vere made against disease c l a s s i f i c a t i o n s , etc. Taking  a l l the above into consideration,  the r e p r o d u c i b i l i t y  vas estimated i n averaging over 25 cases, to be R The  =  19.0  12.7 % .  other method of estimating R i s to replace a(x,t) i n the  above formula v i t h the calculated values of D or PR3,  or any  calcu-  lated c h a r a c t e r i s t i c s . The conditions that the numbers must be from t > 55 minutes p . i . and v i t h i n 3 minutes of one another s t i l l hold. The number found by replacing v i t h PR3 by CRP vas R The alluded  last  =  15.8  14.2  7. .  parameter considered vas the a x i a l  to i n Section 2.1.  resolution  (AR)  I t vas here defined as the r a t i o of the  concentrations at 3mm to the CR peak of the bolus scan after the pre-  107  i n j e c t i o n scan had been subtracted. This d e f i n i t i o n allowed only CRPaligned p r o f i l e s to be used.  Also, diseased eyes were excluded owing  to possible bolus effects. Note that not a l l bolus scans were made at exactly  3 minutes p . i .  Depending on the subjects,  made between 2 and 7 minutes p. i . Averaged  over the three normals, AR  =  0.032 •/- 0.026 .  108  bolus scans were  C Q N C L U S I O N  V I .  The performance of the assembled vitreous fluorophotometer in  close  agreement  with the data that describe the type  detection system used i n t h i s study.  of  Hardware and software  was light  designed  to i n t e r f a c e the l i g h t detection system with a microcomputer provided the s i g n a l conversion, The dence  inherent  on  chamber.  a  data analysis and storage  capabilities.  l i m i t a t i o n of the o p t i c a l system was the depen-  plano-concave contact lens f o r  scanning  the  vitreous  Discomfort caused by t h i s lens from sequential measurements  was l a r g e l y overcome by using a combination of a bandage s o f t contact lens  and  used  by other investigators.  was  a p l a s t i c Luma" lens i n place of the glass  Goldmann  lens  The e f f e c t of substituting the  lenses  a d i f f e r e n t set of F-numbers which were re-calculated by  measu-  ring intraocular distances using The points  most outstanding  along  ultrasound.  problem was locating the same  each scan f o r the purpose of alignment,  data and subtraction i n the subsequent analysis.  reference  reduction  of  The method used  to  lessen the e f f e c t s of alignment errors was to average about an i n t e r val  along  comparison. determined  the p r o f i l e instead of s e l e c t i n g a  specific  point f o r  The effectiveness of t h i s algorithm however could not be because  precise i n vivo data on  vitreous  concentration  could not be obtained. The  correction algorithm f o r the choriod-retinal bolus t a i l i n g  e f f e c t s used i n the Fluorotron" Master d i d not improve the separation between  groups  when applied to the fluorophotometer  109  assembled f o r  t h i s study. The to  penetration r a t i o s i n the 15 diabetic subjects were  increase progressively with the severity of retinopathy  dings),  in  agreement  with published reports.  However,  found  (3 grathere  was  s i g n i f i c a n t dispersion of r e s u l t s about the averages i n each group. In the sample of 16 multiple s c l e r o s i s subjects,  the  penetra-  tion  r a t i o s were not s i g n i f i c a n t l y d i f f e r e n t between the two  that  represented  groups of  standard c l i n i c a l a c t i v i t y categories or  that represented current a c t i v i t y categories.  vitreous fluorophotometry  central the  the  normal  periphlebitis  two  The usefulness  as a non-invasive test  for  monitoring  nervous system a c t i v i t y could not be ascertained because  small sample sizes.  with  groups  of  Abnormal leakage was found i n 4 of 15 cases  vitreous and either minimal or no evidence a c t i v i t y (2).  of  retinal  The penetration r a t i o s i n active  peri-  p h l e b i t i s were elevated (3-4 times normal control).  Abnormal leakage  in  previously  been  elevated penetration r a t i o vas also found i n one  case  the  absence  recorded.  An  of active p e r i p h l e b i t i s  has  not  with vitreous liquefaction. Almost retinopathy  a l l subjects i n the diabetes sample, severity,  irrespective  showed vitreous d i f f u s i o n constants  cantly  less  water.  In the multiple s c l e r o s i s sample and controls,  constants rent.  than the d i f f u s i o n constant of  sodium  of  signifi-  fluorescein i n the d i f f u s i o n  i n the vitreous and i n water were not s i g n i f i c a n t l y d i f f e -  The d i f f u s i o n constant i n the vitreous  was 2-4 times  greater  than the value f o r water i n 4 older multiple s c l e r o s i s subjects and 5 times greater i n the case with vitreous l i q u e f a c t i o n . 110  A P P E N D I X  A  C O M P U T E R  A.1  P R O G R A M M E S  DAS.PRN This prograaae, DAS configures the PIA i n a t r i g g e r i n g node of the Data A c q u i s i t i o n System. The c o n t r o l l i n e s are needed f o r INTR, WR, MUX, s v i t c h A and S/H. RD and CS are held lov, i . e . the output of the ADC i s alvays enabled. This routine, aa entered i n the HBASIC DATA statements, i s separated by a seni-colon and a nunber as seen belov. E.g. - ;#20 This version vas v r i t t e n i n J u l y 1985 by PANG Kian Tiong. PA0-7 to be inputs. - l i n e #1-8 CA1 * not used. - l i n e 419 CA2 = not used. - l i n e #17 • • " » CRA vord * 0010 1113 PB8 PB1 PB2 PB3  1 1 9 ?  to set PA0-7 as inputs. f o r PB6, 7 t o be inputs. f o r PBS, 5 to be outputs. - WR c o n t r o l : - l i n e 49 0 « s t a r t conversion reset and v a i t 1 DUX address: - l i n e #21 PB4 pod * radiometer PB5 S/H address: - l i n e #11 » saitple ' hold PBS Svitch A i n t e r r u p t input l i n e #15 continue reading v a i t t o s t a r t or stop 1 PB7 * ? - INTR i n t e r r u p t input - l i n e #13 0 » INTR h i 1 » INTR l o • • " » DRB vord: 00?? _?011  s  3  EF08 2901 2903 2900 2902 EFDE  »  » » * *  1 lagpos era equ crb equ dra equ drb equ status equ  equ 02901h 02903h 02900h 02902h 0efdeh  0ef08h ;raa/roa junp vector ;control register A ;control register 3 ;data/direction register A ;data/direction register 3 ;PIA s t a t u s  org  0dld6h  ; s t a r t i n g address  r  D1D6  Configuring the PIA begins.  *t  D1DS D1D7 D1DA D1DC  AF 3ADEEF FE02  «•ode:  ca  D1DD 3E2A D1DF 320129 D1E2 3E00 D1E4 320029 D1E7 3E2E D1E9 320129 D1EC 3E00 D1EE 320329 D1F1 3E3F D1F3 320229 D1FS 3E3S D1F8 320329  ;#2  #3  #4  ;#5  xra Ida cpi rz  a  •vi sta nvi  a,02ah era a.0  sta •vi sta  dra a. 02eh era  ;0010.1110 ;port A i s input  nvi sta •vi  a.0 crb a, 03fh  ;0011_U11  sta •vi sta  drb a, 03Sh crb  ;0011_0U0 ;port B i s s e t  status 2  ; i f already i n input mode ;0010_1010  ;#6  111  D1FB 3EB2 D1FD 32DEEF D2B0 C9  0231 3A0229 D234 17 D235 D231D2 0238 3A3229 D23B 17 D23C DA38D2  mvi ata ret  a, 2 atatua  ; Reading and i n t e r r u p t s t a t u a : INTR ;*7 input: Ida drb ral Jnc input ; i f INTR h i  ;*a  high: ;#9  D23F 3A3329 0212 2F D213 77  Ida ral Jc  drb high  Ida cna sov  dra  avi ata ret  a,32bh drb  iii  INTR l o  fl.a  ;#13 0214 3E2B D21S 323229 0219 C9  D21A D21B 021D D21F  F3 0330 3E03 32B8EF  0222 32D3D1 0225 CDD6D1 0228 D22B D22C 0220  3A3229 17 17 DA2302  D233 C33BD2  0233 0234 0236 0233  F3 D3B9 3EB3 3238EF  D23B 3E1B D23D 323229  0243 21D401 D243 3E33 D245 323229 0248 CD31D2  D24B 21D2D1 D24E 3E23 0253 323229 0253 CD31D2  0256 D259 D25A D25B  3A3229 17 17 D263D2  D25E 3EF3 0263 320301 0263 0265 0267 026A 0263  D331 3E31 3238EF FB C9  ;a«13.1311  j F i r a t entry point from BASIC programme. ;#U entry: d i out 3 a, 3 mvi f l a g p o s ; i n Bank 2 ata ;#12 3dld3h jstop code ata mode ;teat PIA atatua call ;#13 a t a r t : Ida ral ral ;poll Svitch A start ;#14 read ; Subaequent e n t r i e s from BASIC r o u t i n e . ;#15 di out mvi a, 3 ata flagpoa ; i n Bank 2 ;#16 ;3331 1311 read: mvi a,31bh eta drb Reading radiometer #17 Ixi h,3dld4h mvi a,333h ;3811_3B11 ;#18 sta drb call input Reading pod. #19 lxi mvi ;#23 ata call  h,3dld2h a,fl23h ;0313_3311 drb input  : Check f o r atop a c q u i s i t i o n ;#21 Ida drb ral ral jnc goback ;#22 a 3f3h ;stop code mvi ata 3dld3h ;#23 out 1 goback: a, 1 mvi flagpoa sta ;back i n Bank 1 ei ;to BASIC r o u t i n e ret r  112  A.2  SCANMENU.BAS 133 110 123 130 143 150 163 170 183 190 233  WIDTH 52 : CLEAR. 4HD1CF : RE.1 Updated 230186 OEFSTR A DEFINT I-N A=" : PRINT A : PRIHT "SCAN MENU" : PRINT A : PRINT RESET : PRINT "(0) Run VITREOUS SCANNING programme" : PRINT PRINT "(1> Run PLASMA SCANNING programme" : PRINT PRINT "(2) Run SUBJECT DATA ENTRY programme" : PRINT PRINT : PRINT "Enter ANY OTHER number to EXIT." : PRINT PRINT "Union programme do you » i 3 h to run *; : INPUT II IF (II>2) THEN END ELSE PRINT IF (11 = 2) GOTO 523  210 PRINT "Loading assembly-language subroutine, DAS. " 223 DATA 4HAF, 1H3A.4HDE.4HEF,4HFE.2, 4HC8 233 DATA 4H3E.lH2A.iH32.1..4H29, AH3E, 3 240 DATA 4H32.3,4H29.AH3E. 4H2E. SH32. i . 1H29 253 DATA 4H3E.3.4K32.3.4K29, SH3E. 4H3F 263 DATA 4H32. 2.SH29.SH3E.AH36.4H32.3,4H29 270 DATA 4H3E.2.AH32.4HDE.1HEF, 4HC9 283 DATA 4H3A.2,4H29.4H17, 4HD2. 1, 4HD2 293 DATA 4H3A,2.4H29,4H17, 4HDA, 3, 4KD2 333 DATA 4H3A.3,4H29,4H2F,4H77 313 DATA 4K3E. 1H23,4H32.2,4H29, 4HC9 323 DATA 4HF3. AHD3.3, AH3E. 3, 4H32.3, AHEF 333 DATA 1H32,4HD3,1HD1.4HCD. AHD6,AHD1 340 DATA 4H3A.2,4K29,4H17.1H17, 4HDA, 4H28. AHD2 353 DATA 4HC3.AH3B. 4HD2 363 DATA 4HF3.4HD3,3.4H3E. 3. 4H32, 3, 4HEF 373 DATA 4H3E.4H13.4H32, 2, 4H2S 330 DATA 4H21.4HD4,4HD1.4H3E. 4H33 393 DATA 4H32,2,4H29,4HCD, 1, 4HD2 430 DATA 4K21,4HD2.4KD1,4H3E. 1K23 413 DATA 4H32.2.4H29,4HCD,1, 4HD2 420 DATA 4H3A.2.4H29. 4H17. 4H17, 4HD2, 4H63. 4HD2 433 DATA 4H3E,4HF3. 4H32.4HD3. 4HD1 443 DATA 4HD3, 1, 4H3E, 1. 4H32, 3. 4HEF, 4HF3, 4HC9, 3, 3, 3 453 FOR 1=1 TO 151 463 READ J : K=4HD1D5»I : POKE K, J 473 NEXT I 480 PRINT : PRINT •=> DAS has been a u c c e s a f u l l y loaded!" : PRINT 493 IF (11 = 1) GOTO 510 500 CHAIN MERGE "B:VITSCAN". 1013, ALL . 510 CHAIN MERGE "B:PLASCAN\ 1313. ALL 523 DIM AA(22),X(3) 533 A="-PRIHT : PRINT A 543 PRINT 'SUBJECT DATA" : PRINT A : PRINT S53 PRINT "Enter the f o l l o v i n g — > • : PRINT 563 LINE INPUT "Subject's NAME -> ";AA(01 : PRINT 573 LINE INPUT "Subject's AGE --> *;AA <1) : PRINT 580 LINE INPUT ' Scan EYE --> *;AA(2) : PRINT 590 LINE INPUT • Sca,n DATE -> ":AA(3) : PRINT : PRINT 603 FOR 1=3 TO 5 STEP 3 613 PRINT 'Enter the "; 620 IF (I<>0) GOTO 643 633 PRINT 'LENGTHS noted ( i n Osb/DAS u n i t s ) —->• : GOTO 663 643 PRINT "ULTRA-SOUND scan r e s u l t s ( i n mm) --->• : PRINT 653 PRINT TAB(13);'If NO ultra-sound taken, enter 3." 663 PRINT : PRINT "VITREOUS ---> "; : INPUT X(I) 673 PRINT : PRINT • LENS ---> •; : INPUT X ( I - l ) 683 PRINT : PRINT • AOUEGUS ---> "; : INPUT XU-2) 693 PRINT : PRINT : PRINT : PRINT 733 NEXT I 710 PRINT 'Amount of FLUORESCEIN i n j e c t e d was •; 723 INPUT X(6) : PRINT : PRINT : PRINT : PRINT 733 PRINT "Enter any REMARKS, COMMENTS or OBSERVATIONS -->' : PRINT 743 PRINT TAB(10);"Press RETURN to e x i t . " : PRINT : 1=3 753 LINE INPUT "Enter -> ';A 763 IF <A=") GOTO 783 773 1=1*1 : AA(I1=A : GOTO 753 783 PRINT : PRINT : PRINT 'The above i s to be f i l e d i n " 793 PRINT : PRINT TAB!13);"A f o r the LEFT D r i v e ' 833 PRINT TAB(13);'B f o r the RIGHT Drive "; 813 A=INPUTSU) : PRINT : PRINT : KEY = 0 820 IF <A="A" OR A="a" OR A="3' OR A="b') THEN KEY=1 833 IF (KEY=0> GOTO 980 340 IF A=-'b" THEN A="B" 853 IF A="a" THEN A="A" :  113  OPEN "a\U,A'":SU3JECT.DAT" : J=0 : !f = 3 : N=3 FOR K = J TO ,1 IF N=l THEN PRINT ELSE PRINT *1,AA(K) NEXT K N=N-1 ON N GOTO 923.930.940 M=6 : GOTO 370 J=4 : »=I : GOTO 373 PRINT : PRINT 'SUBJECT.DA? baa been completed.* ERASE AA.X PRINT : PRINT : PRINT : PRINT 'Da NOT f o r g e t to COPY onto a run Diskette before BATCHRUN !!" 970 PRINT : PRINT : GOTO 130 983 PRINT : PRINT : PRINT 'BAD Entry !!! Please t r y again.* 993 PRINT : PRINT : GOTO 733 860 370 383 390 900 913 920 933 943 953 960  A . 3  V I T S C A N . B A S 1030 1013 1020 1030 1040 1050 1060 1073 1080 1090 1100 1110 1120 1133 1140 1150 1163 1170 1183 1190 1203 1213 1220 1230 1243 1253 1263 1270 1280 1290 1300 1310 1320 1330 1343 1350 1360 1370 1380 1390 1400 1413 1423 1430 1440 1450 1460 1473 1488 1498 1530  REM Updated 343286 DIN IXU633), IY( 1603), NL( 9) A="*»»-'">**->-*"»'-*" : PRINT : PRINT : PRINT A PRINT "VITREOUS SCANNING' : PRINT A : PRINT : PRINT PRINT "SYSTEMS CHECKS" : PRINT PRINT '(1) A l l equipment ON.' : PRINT '(2) A l l c a b l e s connected." PRINT '(3) S»itch A LOW. • : PRINT '(4) VOLTMETER/Intensity d e t e c t o r ON." PRINT '(5) I n t e n s i t y = 141 •/- 1" : PRINT '(6) HV = 739" PRINT '(7) R/H exp = 0" : PRINT '<8) R/H-DAS output » 3' PRINT '(9) Switch A HIGH to begin." : PRINT K=&HD21A : IP=1HD233 : PRINT : PRINT "LANDMARK SCANNING" PRINT : CALL K : A=INKEYS : J«=-l CALL IP : PRINT " I n t e n s i t y =•;PEEKf1HD1D4); K=PEEX1&HD1D2> : PRINT TABI30);"Pod =';K IF (INKEYS*'") GOTO 1163 JM=jn*l : 1XIJH>=K : PRINT CHRSK7); IF (PEEK(4HD1D3)<>243) GOTO 1120 IF J(1<1 THEN GOTO 1250 ELSE PRINT IF JH>8 THEN J(l = 8 FOR K«I TO JH I = K-1 : HL(I)»IXCK)-IXU> PRINT TAB<18>;'Hark';K;'- Nark';I;'*•;BL(I) NEXT K PRINT : PRINT " T o t a l LENGTH '';IX(JN)-IX(3) : PRINT PRINT "IMPORTANT: WRITE do»n these numberB.' : PRINT PRINT : PRINT "LANDHARKing »ill NOT be Repeated, OK?" : GOSUB 1738 IF <A»CHRS<27>) GOTO 1100 A*1NKEY9 : PRINT : PRINT "Which EYE i a to be scanned?" > PRINT PRINT ' Enter : ESC f o r the RIGHT eye, • PRINT TAB(16)j"ANY f o r the LEFT eye. "; : A=INPUTS(1) : PRINT : PRINT IF A=CHRS<27> THEN JJ=1 ELSE JJ=0 IF JJ=0 THEN PRINT 'LEFT"; ELSE PRINT "RIGHT"; PRINT " Eye . i l l be scanned.• : PRINT : PRINT PRINT "SCANNING INSTRUCTIONS" : PRINT " = = = = = » = = = = .. = " = » = = = = =• PRINT '(1) Scan STEADILY (from the RETINA each t i m e ) . " : PRINT PRINT "(2) Wait f o r the BEEP before s t a r t i n g . " : PRINT PRINT "(3) S»itch A t o HIGH t o BEGIN scanning." : PRINT K=&HD21A : IP=4HD233 : CALL X FOR 1 = 0 TO 123 CALL IP : IM=PEEK < &HD1D2) IF K 5 0 THEN PRINT "Pod *";IN, "Retina =" ;PEEK ( &HD1D4) IF 1=50 THEN PRINT CHRS<26> NEXT I I=-l : PRINT CHRS(7)j IF 1-1599 THEN GOTO 1470 ELSE I = I»1 CALL IP : IX(I)= P£EK< &HD1D2) : IY(I)=PEEK(SHD1D4) IF (PEEK<&HD1D8><>248> GOTO 1440 ELSE GOTO 1488 PRINT "You are out o f MEMORY. S»itch A to LOW.• t PRINT N-I : GOSUB 1778 : PRINT : PRINT IF (KO<>8) GOTO 1598 PRINT " l a the scan t o be SAVEd?" : GOSUB 1738 : PRINT  114  1510 1520 1530 1540 1550 1560 1S70 15B0 1590 1600 1610 1620 1630 1640 1650 1660 1670 1680 1690 1700 1710 1720 1730 1740 1750 1760 1770 1780 1790 1800 1810 1820 1830 1B40  IF A=CHR9(27> THEN PRINT "NO I" E L S E PRINT "YES I" PRINT : PRINT 'Please CONFIRM v i t h the SAME • : GOSUB 1730 : PRINT IF A=CHRS(27> THEN GOTO 1590 ELSE PRINT PRINT "Enter the P.I. Scan TIJ1E" : PRINT TAB(20);"for the •; IF J J = 0 THEN PRINT "LEFT"; ELSE PRINT "RIGHT"; LINE INPUT " eye. ";A : PRINT : PRINT NAHE AA*"TEH.DAT* AS AA*A*".DAT" : PRINT PRINT "===>> The f i l e has been s u c c e s s f u l l y RENAHEd. " PRINT : PRINT "Is SCANNING t o be CONTINUEd?" : GOSUB 1730 IF (A=CHRS<27)) GOTO 1630 PRINT "SYSTEMS .111 NOT be checked, OK?" : GOSUB 1730 IF <A«CHRS<27>> GOTO 1270 E L S E GOTO 1040 PRINT : PRINT : PRINT "The f l l e a i n Drive A a r e :" PRINT : FILES • A: *.• • : PRINT : PRINT PRINT "The f i l e s i n Drive B are : • : PRINT FILES " 8 : : PRINT : PRINT PRIHT "Do you v i s h t o r e t u r n to the SCAN MENU?" : GOSUB 1730 IF <A=CHR3<27)> GOTO 1590 ERASE IX,IY,ML ON ERROR GOTO 0 CLOSE : GOTO 130 REM Prompt o p t i o n s d i s p l a y . A=INKEYS : PRINT : PRINT TABI5);"Enter : ANY KEY f o r YES" PRINT TAB<17)j"ESC f o r NO "; : A=INPUTS<1> PRINT : PRINT : PRINT : RETURN REM Subroutine f o r v i d e o - p l o t t i n g . WIDTH 127 K=155 : W=0 : O=20!/<K-W> : IR=IX(N) : LL=IX<0>-10 : Kg=0 I F LL<0 THEN LL=0 IF ( I R - L L X 5 0 THEN U=ll ELSE U=1251 / (IR-LL) PRINT CHRS(26);CHR9(27)•")•;CHRS(27)•" = •-CHRS(54)«CHRS(36) ; PRINT " Press ";CHRS(27)••(";"ANY";CHRS<27)•">"; PRINT " KEY t o ABORT t h i s SCAN at any time.• : A=INKEY9 FOR 1=0 TO 20  1850 1860 1870 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 2060 2070 2080 2090 2100 2110 2120 2130 2140 2150 2160 2170 2180 2190 2200 2210 2220 2230 2240 2250  PRINT CHR9<27).'=".CHRS<32.I>>CHRS<32>; IF (1=0 OR 1=5 OR 1=10 OR 1=15 OR 1=20) THEN PRINT "«•; ELSE PRINT " I " ; NEXT I FOR 1=0 TO 12S PRINT CHRS(27>."=-"<-CHRS(53)-CHRS(33.I) ; IF (IHKEYSo"" ) GOTO 2180 IF RIGHTSISTRSII),1>="0" THEN PRINT ••»•; ELSE PRINT "-"; NEXT I IF JJ = 0 THEN AA= "A: " ELSE AA="B:" OPEN "0",I3, AA*"TEH.DAT" : PRINT 13, IH : A=INKEY9 PRIHT CHRS(27)•" = "*CHRS(54 > *CHRS(32);LL; : H»0 PRINT CHR9(27)»"-"»CHSS(54).CHR9(150)jIR; FOR 1=0 TO N IF I I H K E Y 3 0 " " ) GOTO 2210 IF (IY(I) = 128 OR IY(I) = 127) GOTO 2050 IP=CINT((IX(I)-LL).U) : K=CINT( (IY(I)-W).0) PRINT »3,IXtl) : PRINT »3,IY<I> : K-H'l IF (1P>124 OR K>20) GOTO 2050 IF (IP<0 OR K<0) GOTO 2050 PRINT CHR9(27)."»".CHR9(52-K).CHR9(33»IP);"." NEXT I IP=IM-LL : K»CINT(IP"U) : PRINT CHR9(27)••("; PRINT CHRS<27)."="«CHRS<34>>CHRS(35>iM;"pairs were e n t e r e d . " PRINT CHRS ( 27) • " = " +CHR3 (53) *CHRS(33*K) ; "R" ; IF (JH<0> GOTO 2150 FOR 1=0 TO JM I P = I P - H L U ) : K=CINT(IP«U) IF <K<0 OR K>124) GOTO 2140 PRINT CHRS(27)."=".CHR9(53)«CHR9(33»K);"!"; NEXT I A=INKEYS : PRINT CHRS(27)•"="*CHRS(54)*CHRS(40)f PRINT "Press ANY key t o r e t u r n to PROMPT mode."; : A=INPUT9(1) CLOSE PRINT : PRINT : PRIHT CHRS<27>«"<" : PRINT WIDTH 52 RETURN PRINT CHR9(27)."=".CHR9(35)'CHRS(35); PRINT "Please CONFIRM that you vant t h i s scan ABORTEDl I " GOSUB 1730 : PRINT : ATT=INKEY9 : ATT=INKEY9 IF <A=CHR9(27)> GOTO 1990 K0=1 : GOTO 2170  115  A.4  P L A S C A N . B A S 1000 1010 1020 1030 1040 1050 1060 1070 1080 1090 1100 1110 1120 1130 1140 1150 1160 1170 1180 1190 1200 1210 1220 1230 1240 1250 1260 1270 1280 1290 1300 1310 1320 1330 1340 1350 1360 1370 1380 1390 1400 1410 1420 1430 1440 1450 1460 1470 1480 1490 1500 1510 1520 1530 1540 1550 1560 1570 1580 1590 1600 1610 1620 1630 1640 1650 1660 1670 1680 1690 1700 1710 1720 1730 1740  REM Updated 180186 DIM IXI1200). IY(1200),L(256), V( 256), S( 256), X(55), Y(55),Z(55) A="«->*»'<-»**»»»**»' : PRINT : PRINT A : KEY=-1 PRINT 'PLASMA SCANNING" : PRINT A : PRINT : PRINT PRINT : PRINT 'SCANNING INSTRUCTIONS" : PRINT ' = = = . = = = « » = . . « = = . . . ' PRINT "(1) Aim PROBE with WHITE l i g h t . " ! PRINT PRINT "(2) Adjust c e l l l o r NO REFLECTION.' : PRINT PRINT "131 Move PROBE u n t i l J u s t behind g l a s s s u r f a c e . ' s PRINT PRINT "(4) F i x the POD p o s i t i o n . " : PRINT PRINT '(5) S«itch A to HIGH to BEGIN s c a n n i n g . ' K=IHD21A : IP=1HD233 : CALL X : PRINT CHRSI26); FOR 1=0 TO 99 CALL IP : K=PEEK(&HD1D2) : K=PEEK(&HD1D4) NEXT I I=-l : PRINT CHRS(7); IF 1 = 1199 THEN GOTO 1180 ELSE I-I»l CALL IP : IX (I) =PEEK ( &HD1D2) s IY (I) =PEEK (&HD1D4) IF <PEEK<SHD1D0><>240) GOTO 1150 ELSE GOTO 1190 PRINT "You are out of MEMORY. S v i t c h A t o LOW. " : PRINT NN=I : PRINT "AVERAGING begins." : PRINT : PRINT PRINT TAB (5); "Press ANY key t o i n t e r r u p t . " : PRINT : PRINT FOR J=0 TO 255 L(J)=0 : V(J)=0l : S(J)=0l : A=INKEYS NEXT J FOR J=0 TO NN I=IX(J) : HM=IY(J) IF (INKEYSo"") GOTO 1450 IF (MM=127 OR MM=128) GOTO 1290 L(I)=L(I)«1 : V(I)=V(I).MM : S(I)=S(I)*MH*2 NEXT J J=-l : JM=0 : KS=0 FOR 1=0 TO 255 MM=L(I) : KS=XS*MM IF (INKEYSo"") GOTO 1450 IF (HM<2) GOTO 1390 J = J»1 : V!J)=V(I)/MH : L(J)=I S<J)=<S(I)-MH«V(J) 2)/<HH-1> IF (MM<JM> GOTO 1390 JH=MH : M=J NEXT I KEY=KEY*1 : Y<KEY)=V(M) ! Z(KEY)=SOR<S<H>> PRINT : PRINT "AverBge ••;V(M)j•«•/-•;Z<KEY) 1 PRINT PRINT TAB(10);"for";JM;"out of» ;KS s " p o i n t s . • : PRINT PRINT : PRINT 'What i s the Sample TIME ( i n min. P.I.) "; INPUT X(KEY) : PRINT : PRINT : GOTO 1460 PRINT TAB(5);"Averaging i n t e r r u p t e d II* : PRINT : PRINT PRINT : PRINT 'Is SCANNING to be CONTINUEd?' : GOSUB 1720 IF <A<>CHRS<27)> GOTO 1040 PRINT : PRINT "Was the scan f o r (0) BOTH eyes 7" PRINT TABC18>;"(1> LEFT only 7* 1 PRIHT TAB(18>;'<2) RIGHT only •; INPUT KM : PRINT : PRIHT IF KM=2 THEN J=l ELSE J=0 A= *:PLASMA. DAT* : PRINT "The PLASMA data a r e : " : PRINT IF J=0 THEN OPEN '0",#2, "A'-A ELSE OPEN "0",»2, "B'-A FOR 1=0 TO KEY PRINT #2,XII) : PRINT <2,Y(I) : PRINT #2,Z(I) PRINT I + l ; ' ) ';X(D, Y d ) : ' •/- ' ; 2 ( D NEXT I CLOSE : J=J-1 : PRINT : PRINT IF (KM=0 AND J=l) GOTO 1530 PRINT "Plasma averages have been f i l e d . • : PRINT PRIHT 'The f i l e s i n Drive A are : * 1 PRIHT : FILES 'A:«.»' PRINT : PRINT : PRINT 'The f i l e s i n D r i v e B are : • PRIHT : FILES 'B: •. • • : PRINT : PRINT : PRINT PRINT "Do you have another s e t o f samples to da?" : GOSUB 1720 IF A=CHRS<27> THEN GOTO 1690 ELSE RESET PRINT : PRINT "Enter ANY key a f t e r changing LOGGED d i s k e t t e . • A=INPUTS(1) : RESET : PRINT : PRINT GOTO 1020 ERASE IX, IY.L, V,S,X,Y,Z PRINT : PRINT : PRINT : PRINT CLOSE : GOTO 130 A'INKEYS : PRIHT : PRINT TABI5)! "Enter : ANY KEY f o r YES" PRINT TABI17)j*ESC f o r NO ' j : A'INPUTSU) : PRINT : PRINT ! PRINT RETURN A  116  A . 5  B A T C H R U N . B A S 100 110 120 130 140 150 160 170 180 198 200 210 220 230 240 250 260 270 280 290 300 310 320 330 340 353 360 370 380 398 400 413 423 430 440 450 460 470 480 490 503 510 520 S30 540 550 560 570 580 590 608 618 620 630 640 650 660 670 680 690 700 710 720 730 740 750 760 770 780 790 833 810 820 830  WIDTH 52 ; RED Updated 120386 DEFINT I-N DEFSTR A DIN AN(13>, ITASKCU) GOTO 170 GOTO 680 RESET PRINT CHRSI26);"Are you using a COPY o f the o r i g i n a l data?" A='-»*»».»-«-»»«->'" : PRINT : PRINT A PRINT "PROGRAMME MENU" : PRINT A : PRINT PRINT "<0> Run REDUCE - Ra» data averaging" PRINT "(1) Run B/G - Background averaging" PRINT "(2) Run MINUS - Background s u b t r a c t i o n " PRINT "(3) Run SUDATA - Reprod. /LLoD/AxRes" PRINT "(4) Run BLOOD - Plasma data i n t e g r a t i o n " PRINT "(5) Run C/VA2 - Cunha-Vaz's a l g o r i t h m " PRINT *<6) Run SLOPES - C u r v e - f i t t i n g method" PRINT "(7) Run PLOT - Coarse p l o t t i n g " PRINT "(8) Run DRAW - Super-impose p l o t t i n g " PRINT "(9) Run LUND - Lund-Andersen's a l g o r i t h m " FOR 1=0 TO 10 ITASK(I)=-99 NEXT I PRINT : PRINT "Enter the SEOUENCE t o run ->" : PRINT PRINT "Runs ALWAYS begins v i t h the l e f t t o r i g h t . • : PRINT PRINT TAB(16);"Yes = ANY key : No = ESC* PRINT : JTASK=-99 : A=INKEY9 : A""" : NTASK=-99 PRINT "0 1 2 3 4 5 6 7 8 9* FOR 1=0 TO 9 A=INPUTS(1) : PRINT * "; IF A=CHR9<27) THEN GOTO 428 ELSE ITASKU>=55 IF I<7 THEN JTASK=5S ELSE NTASK=55 NEXT I IF (JTASK>0 OR NTASK>0> GOTO 490 PRINT : PRINT : PRINT "NO task assigned!* : PRIHT PRINT "Do you «ant t o c o n t i n u e ! <Y> *; A=INPUT9(1) : PRINT : PRINT IF (A="N* OR A="n") THEN LPRINT CHR9C27);CHR9(79)J ELSE GOTO 160 END PRINT FOR 1=0 TO 9 IF ITASK(I)>0 THEN A="Y* ELSE A="N" PRINT A;* NEXT I PRINT : PRINT : PRINT TAB(15);"Please CONFIRM I (Y) *J : A=INPUT9(1) IF (A="N* OR A="n") GOTO 170 IF (JTASK<0 AND NTASK=55> GOTO 640 PRINT : PRINT : PRINT " F i l e s i n D r i v e B are:* PRINT : FILES "B:*.«" : PRINT : PRIHT : PRINT PRINT TAB!5);"Enter i n CHRONOLOGICAL, ASCENDING order.* PRINT : PRINT TAB<18);"Enter RETURN t o e x i t . * : PRINT : HO--1 LINE INPUT 'Enter --> "jA IF <A=""> GOTO 640 MQ=HO-l : AN< MQ) = A : GOTO 610 PRINT : PRINT "Check PRINTER/PAPER.* : PRINT LPRINT CHRS< 27) ;CHRS (65) ; CHRS (12) ; CHRS ( 27) ;CHR9( 50) ; LPRINT CHR9<18);CHR9(27);CHRS<73);CHR9(1); LPRINT CHRS(27);CHR9(78);CHRS(6); : LTASK=0 NERR=0 : PSCALE'.095919 FOR KTASK=LTASK TO 9 IF ITASK ( KTASK) <3 THEN GOTO 820 ELSE LTASK=KTASK*1 ON KTASK GOTO 733,740,753,760,770,780,790,800,810,820 CHAIN MERGE * A:REDUCE*, 1010, ALL CHAIN MERGE 'A:B/G", 1010, ALL CHAIN MERGE "A:MINUS",1010,ALL CHAIN MERGE "A:SUDATA",1010,ALL CHAIN MERGE "A:BLOOD",1010,ALL CHAIN MERGE "A:C/VA2", 1010,ALL CHAIN MERGE "A:SLOPES", 1010, ALL CHAIN MERGE "A:PLOT*, 1013, ALL CHAIN MERGE "A:DRAW",1010,ALL CHAIN MERGE *A:LUND", 1010, ALL NEXT KTASK PRINT s PRINT : GOTO 450  117  A . e 1000 1010 1020 1030 1040 1050  R E D U C E . B A S REM Updated 210186 A = " « : PRINT A : PRINT •REDUCE" : PRINT A DIN 1(256), IXI4000),IYI4000), IS<256), Y(256),WI256) DEF FNCQNC(X)=EXP(-109.339*SQR(11915.9»X/.114441)) ON ERROR GOTO 1790 A="»  1060 FOR LT=0 TO HO 1070 FOR K=0 TO 255 1080 I(K>=0 : IS(K)=0 1090 NEXT K 1100 JC0=255 : JCR = 0 : OPEN "I",#1,"B:"-AN(LT)•\ DAT" i INPUT #1,N : K = - l 1110 IF E O F I D O 0 THEN GOTO 1160 ELSE K = K-1 1120 INPUT #1,J : INPUT #1,IY(K) : K J X K J X l 1130 IF J>JCR THEN JCR=J 1140 IF J<JCO THEN JC0=J 1150 I X I K X J : GOTO 1110 1160 CLOSE 11 : JVL=0 ; IHA=K ; NUN»-1 : JMV = 0 1170 FOR K=JCO TO JCR 1180 IF ( I I X X 2 ) GOTO 1530 1190 PRINT "For X =";K;", there are" ; K K ) ; " p o i n t s . " : PRINT 1200 IHAX=0 : IHIN=255 : J ' - l : JLE=I(K> : U=0i : V»0l : KP=0 1210 FOR KX=JHV TO IMA 1220 IF (KP=1 OR IX(KXX=K> GOTO 1240 1230 KP=-1 : IMY=KK 1240 IF I X ( K K X > K THEN GOTO 1290 ELSE JAL=IY(KK> 1250 IF J=(JLE-1) THEN GOTO 1300 ELSE J=J-1 1260 IS(JAL) = I S ( J A L X 1 : U=U'JAL : V=V'JAL 2 1270 IF JAL>IMAX THEN IMAX=JAL 1280 IF JAL<IHIN THEN IMIN=JAL 1290 NEXT KK 1300 IF KP=t THEN JHV=IHV ELSE JMV'KK 1310 KK=I(1AX-II1IN : PRIHT "MAX =" i IMAX, "MIN = * ; IMIN, "MAX-NIN = ";KK 1320 IF XK<7 THEN GOTO 1530 ELSE JLE=0 1330 IF JVL=0 THEN P=U/(J*1> 1340 FOR KK=IMIN TO IMAX 1350 IF ( I S I K K X 0 ) GOTO 1430 1360 PRINT " For Y =•;KK;',";TAB(18))"there are";IS(KK>;"points. " 1370 I F ( I S ( K K X J L E ) GOTO 1430 1380 IF IS(KK)>JLE THEN KJ=255 1390 ICO=ABS(CINT(KK-P)) 1400 I F (ICO>KJ) GOTO 1420 1410 KJ'ICO : IHV=KK 1420 JLE=IS(KK) 1430 NEXT K K 1440 KK'IHV : PRINT : PRINT "Estimate used i s " ; K K i V-0t t U-01 l JLE-9 1450 ICO=KK»5 : ICR=KK-5 : KK=CINT<KK-P> 1460 IF KK>5 THEH ICR=CINT(P) 1470 IF KK<-5 THEN ICO=CINT(P> 1480 FOR KK=IHIN TO IHAX 1490 IF <KK>ICO OR KK<ICR) GOTO 1510 1500 JLE=JLE.IS(KK) : V»V-IS(KK>»KK 2 i U=U*KK"IS(KK) 1510 NEXT KK 1520 PRINT : PRINT "Upper Bound *•;ICO, "Lover Bound =";ICR 1530 U=U/JLE : PRINT : PRINT TAB(10);"No. o f p o i n t s used - • ; J L E ; " o f • ; I ( K ) 1540 FOR KK=IHIN TO IMAX 1550 IS(KK)=0 1560 NEXT KK 1570 IF JLE<2 THEN GOTO 1630 ELSE V=(V-JLE«U 2>/(JLE-1> 1580 IF V=0l THEN V= 11 1590 IF V<0l THEN V=ABS(V) 1600 NUM=NUM'l : V=SQR(Y) : HNUMXK ; Y(NUH)=U : W(NUM)=V 1610 PRINT : PRINT •(•;AN(LT);•) Ans i";K,UJ•-/-•;V 1620 JVL=1 : P=U : PRIHT A 1630 HEXT K 1640 OPEN "0",#2. "Br"*AH(LT)*".AVG" : PRINT " F i l i n g ";AN(LT)J". AVG" 1650 FOR K=0 TO NUN 1660 KK=I(K)-N : U=FNCONC(Y(K)) : V'FNCONC<W(K>) 1670 PRINT #2, KK : PRINT 12, U : PRINT 12, V 1680 NEXT X 1690 CLOSE : AA="B:•'AN(LT)•". DAT" 1700 NAME AA AS "B:TEM. TEH" 1710 PRINT : PRINT AN(LT)j". DAT has been renamed." 1720 KILL "B:TEM.TEM" 1730 PRINT : PRINT AN(LT);". DAT has been eraBed.• ! PRINT A 1740 NEXT LT 1750 ERASE I,IX,IY,IS,Y,W A  A  A  118  1769 ON ERROR GOTO 9 1770 CLOSE : GOTO 150 1780 REN E r r o r subroutine to p r i n t p o s i t i o n s vhere e r r o r s occur. 1790 IF NERRM0 THEN END ELSE NERR=N£RR*1 1800 LPRINT "REDUCE E r r o r Code «";ERR;"in Loop * " ; L T j " a t L i n e #";ERL 1810 RESUME NEXT  •7  B / G .  B A S  1000 REM Updated 210186 1010 DIN X(3,256),Y(3,256),Z(3,256),L(4),H<256),C(256>,D<256> 1020 PRINT : PRINT "B/G" : PRINT •••»• : PRINT : NN=-1 1030 FOR 1=0 TO MO 1040 IF VAL(AN<I))=0 THEN NN=NN*1 1050 NEXT I 1060 IF (NN=0) GOTO 1570 1070 FOR 1=0 TO 255 1080 M(I)=0 : C(I)=0! : D(I)=0l 1090 NEXT I 1100 ON ERROR GOTO 1610 : XHIN=2551 1110 FOR N=0 TO NN 1120 ON N GOTO 1140,1150 1130 A="0" : GOTO 1160 1140 A="00" : GOTO 1160 1150 A="000" 1160 OPEN "I",#1,"B:"'A'".AVG" : I=-l 1170 PRINT : PRINT TAB(10);"Reading ";A;".AYG" : PRINT 1180 1=1*1 : INPUT #1,X(N,I) : INPUT #1,Y(N,I> : INPUT «1,Z(N,I> 1190 IF X(N, IXXMIN THEN XMIN = X(N, I) 1200 IF <EOF(1)=0) GOTO 1180 1210 CLOSE : L(N)=I 1220 NEXT N 1230 PRINT : XMIH=ABS(XMIN) : PRINT TAB!10);"Min =";XMIN : PRINT 1240 FOR 1=0 TO N-l 1250 FOR J=0 TO L(I) 1260 K=CINT(X<I, Jl'XMIN) : C ( K ) = C ( K ) * Y ( I , J ) 1270 D(X)=D(K)»Z(I, J)"2 : H<K)=M(K)*1 1280 NEXT J 1290 NEXT I 1300 K = - l 1310 FOR 1=0 TO 255 1320 IF H(I)=0 THEN GOTO 1350 ELSE K = K-1 1339 IF H<I>=1 THEN J=2 ELSE J=MII) 1340 C(K)=C(I)/M<I) : D ( K ) = S O R ( D ( I ) / ( J - D ) : M(K)-I-XBIH 1350 HEXT I 1360 OPEH "O",12,"B:B/G.DAT" 1370 FOR 1=0 TO K 1380 PRINT #2,M(I) I PRINT #2,C(I) : PRINT #2,D(I) 1390 NEXT I 1400 CLOSE 1410 KILL "B:0.AVG" 1420 NAME "B:B/G. DAT" AS "B:0.AVG" 1430 FOR 1=1 TO NN 1440 ON I GOTO 1450,1470,1490 1450 KILL "B:00.AVG" 1460 GOTO 1500 1470 KILL "B:000. AVG" 1480 GOTO 1500 1490 KILL "B:0000.AVG" 1500 NEXT I 1510 N=0 : AN(N)="0" 1520 FOR 1=1 TO MO 1530 IF (VAL(AN(I))=0) GOTO 1550 1540 N=N*1 : AN(N)=AN(I) 1550 NEXT I 1560 MO=H 1570 ERASE X, Y, Z, L, H, C, D 1580 PRINT "B/G - average completed and f i l e d . " : PRINT 1590 ON ERROR GOTO 0 1600 CLOSE : GOTO 150 1610 IF (ERL»1160 AND ERR=53) THEN RESUME 1230 1620 IF NERR>10 THEN END ELSE NERR=HERR*1 1630 LPRINT "B/G ERROR Code #";ERR;" i n L i n e #";ERL 1640 RESUME HEXT  119  A .  a  M I N U S . B A S 1000 REM Updated 210186 1010 DIM X(2,256),Y(2.256),Z(2.256),M(2),W<2) 1020 A="-*-.." : PRINT A : PRINT "MINUS" : PRINT A : PRINT 1030 ON ERROR GOTO 1350 1040 FOR LT=0 TO HQ. 1050 IF LT=0 THEN 1=0 ELSE 1*1 1060 OPEN "I\#1,"B:*-AN(LT)-".AVG" : K»-l : U = 0! 1070 K=K«1 : INPUT #1,X(I,K> : INPUT #1,Y(I,K) 1080 INPUT #1,P : ZiI,K)=P 2 1090 IF <X(I,K>>20 OR Y U . K X U ) GOTO 1110 1100 U=Y(I,K) : W<I>=XU,X> 1110 IF EOF(1)<>0 THEN CLOSE ELSE GOTO 1070 1120 H<I)=K 1130 IF (LT=0) GOTO 1300 1140 J=0 : ICR=3 : ICO=0 1150 IF J=8 THEN A=". RET" ELSE A=".CRP" 1160 OPEN "0",#2, "3:"*AN(LT)-A : JCO'-l : JAL=0 : JVL=X (0, M (0))-ICO 1170 FOR K=0 TO M(I) 1180 JMV=X<I,X)-ICR 1190 IF (JMV>JVL) GOTO 1270 1200 FOR JCR=JAL TO M(0) 1210 JLE=X(0,JCR1-ICO 1220 IF JMVoJLE THEN GOTO 1250 ELSE JCO=JC0*l 1230 U=Y(I,K)-Y(0,JCR) : V=S0R(Z(I, X)*Z(0,JCR) ) : P'JMV 1240 PRINT #2,P : PRINT #2, U : PRINT #2, V : JAL=JCR-1 : GOTO 1260 1250 NEXT JCR 1260 NEXT K 1270 PRINT : PRINT TAB( 10) ;AH(LT) <-A; has"; JC0*1; "data-seta. " 1280 CLOSE : J=J-1 : ICR=W(I) : ICO=W<0> 1290 I F (J=l) GOTO 1150 1300 NEXT LT 1310 ERASE X, Y,Z,N,W 1320 ON ERROR GOTO 0 : PRINT : PRINT 1330 CLOSE : GOTO 150 1340 REM E r r o r subroutine to p r i n t p o s i t i o n s where e r r o r s occur. 1350 IF NERRM0 THEN END ELSE HERR=NERR»1 1360 LPRINT "MINUS ERROR Code #";ERR;"in Loop #";LT;"at L i n e =";ERL 1370 RESUME NEXT A  1  A . 9  S U D A T A . B A S 1000 1010 1020 1030 1040 10S0 1060 1070 1080 1090 1100 1110 1123 1130 1143 1153 1163 1170 1180 1193 1233 1213 1220 1230 1240 1250  REM Updated 070386 DIM XN(3),B(3),C<3) A="--.-.." AA="SUBJECT DATA" : PRINT PRINT A : PRINT AA : PRINT A : PRINT LPRINT SPC(27)CHRS(14);AA : LPRINT : LPRINT OPEN "I",#1."3:SUaJECT.DAT" : LPRINT : LPRINT INPUT #1.AM : LPRINT S?C(30)"Name : ";AM : LPRINT INPUT >1,A : LPRINT S?C(30)" Age : ";A : LPRINT INPUT #l,A : LPRINT SPC(30)" Eye : ";A : LPRINT INPUT #1.A : LPRINT SPC(30)"Date : ";A : LPRINT : LPRINT INPUT #1,XN13) : INPUT #1.AN(1) : INPUT #1,XN(2) INPUT #1,3(0) : INPUT #1,B(1) : INPUT #1,3(2) C(3)=3(0)/XN(3) : C! 1)=B111/XN(1) : C(2)=B(2)/XN(2) INPUT #1,00 : aa=2S3!.QQ : LPRINT 3PC(18)"FLUORESCEIN i n j e c t e d " GOSUB 1373 : LPRINT : LPRINT LPRINT SPC!10)"Scaling by Lund-Andersen's F-numbers :" QQ=.13812»XN(0> : GOSUB 1333 : GOSUB 1390 : PP=QH BQ=.15731»XN(1> : GOSUB 1340 : GOSUB 1390 : PP=PP*GO Q0=.13938«XN(2) : GOSUB 13S3 : GOSUB 1390 : QO=PP-GO GOSUB 1363 : GOSUB 1393 : LPRINT : LPRINT LPRINT SPCI13)"Ultra-sound r e s u l t s :• 00=313) : GOSUB 1333 : GOSUB 1390 : PP=QO 00=3(1) : GOSUB 1340 : GOSUB 1393 : PP = PP-CO. 00=3(2) : GOSUB 1353 : GOSUB 1390 : 00=PP-00 GOSUB 1360 : GOSUB 1390 : LPRINT : LPRINT LPRINT SPC(10)"Recalculated F-numbers :• :  120  12S0  O.Q=C(0)/PSCALE  : GQSU3 1330  : LPRINT  1270  SG=C(1)/PSCAL£  : GOSUB 1340  : LPRINT  1280  Q0=C(2)/PSCAL£  : GOSUB 1350  : LPRINT  1290 LPRINT S P C I 1 0 ) "REMARKS Si COMMENTS 1300 I F ( E O F t l ) < > 0 ) GOTO 1310  INPUT  1320 CLOSE  #1,A  : GOTO  1300  : LPRINT C H R S ( 1 8 ) ; C H R S ( 1 2 ) ; : GOTO Chamber";  1340  LPRINT SPC134)"Len3*;  : GOTO  1350  LPRINT S P C ( 2 3 ) " A q u e o u s Chamber";  1360  LPRINT S P C ( 4 1 ) '  1370 LPRINT  1400  : GOTO  1370  1370 : GOTO  1370  • : LPRINT SPCI26) " A x i a l  Length";  " " ; C H R S ( 2 4 7 ) ; " ";  LPRINT USING  1390 LPRINT  : LPRINT  1320  : LPRINT SPC1201A  1330 LPRINT S P C ( 2 2 ) " V i t r e o u a  1380  : LPRINT  :•;CHRS<27);CHRS(58)  " mm"  " ####. ###" ;QQ;  : RETURN  : RETURN  1400 DEF F N S L I J , ? ) = C ( J ' 1 ) « ( P - X N < J > >-3<J) 1410 ON  ERROR GOTO  2770  1420 X N ( 1 ) = X H U ) - X N ( 0 ) 1430  B<1)=3<1)'B(0>  1440  XMV=XN(0)/2  1450  YL=XMV-10!  1460 DIM  1500 1510  : X M A = ( X N ( l ) * X N ( 2 ) ) / 2 : XL=XMV-10! : PRINT  : PRINT  : KEY=0  : LPRINT  AF=".RET"  : LPRINT  : LPRINT  : GOTO  1520  I=KZ  TO  MO  SY(0,I)=-10!  1540  SY(4,I)=SY(2,I)  1550  PP=0l  : ?T=0!  : PD=0!  1560  JJ=-1  : PRINT  TAB(18);"Reading  1570  OPEN  1580  : SY<2,I)=lE-22 : SY(3,I)=-I0!  INPUT  #3,PX  : INPUT  JJ=JJ»1  1600  I F (PX<XL OR  1618  PT=PT-1!  1650 1660 1670 1680 1690 1780 1710 1720 1730 1740 1750 1760 1770 1780  #3,3  : ZP=255! : OT=01  ?X>YL)  GOTO :  SXC1,I)=PX-C!0>  OR  13.R  1620  PZ=PZ-S 2 A  : SY(1,I)=0  : SZ(1,I)=R  : YP=WP  : SY(5,I)=Q : SZ(5.I1=R  : ZP = VP  1680  SX(0,I)=PX'C(0) : SY(0,I)=Q : SZ(0,I)=S I F (PX<0! OR  ?X>XN(0)  OR  Q>SY<2,1)) GOTO  SX(2.I)=PX-C(0) : SY(2.I)=a I F (PX<XMV OR  ?X>XN(1)  I F ?X< = XN(0) THEN SY(3.I)=g  OR  Q < S Y ( 3 , I ) ) GOTO  SX(3.1)=PX-C<8)  PX>XN(2)  OR  I F (PX<XMA OR  Q < S Y ( 6 , I ) ) GOTO  I F PX>XN(0)  THEN  I F (00>3.5 OR OT=QT-l!  1770  GOTO 1830 E L S E 00<2.5) GOTO  : SZ(6,I)=R OQ=C<0)-PX  1S00  : QD=OD-0 : 0Z=QZ-R*2  I F (QO<PP) GOTO  1830  : SX(IM,I)=QO  CLOSE  1850  I F (PT=0!) GOTO  : SYIIM,I)=0  : PP=PP*3!  I F ! E O F ( 3 ) = 0 ) GOTO : PRINT  • done  1580 1"  : PRINT  1900  1860  SX(9,1)=XMV-C(8)  : SY(9, I)=PD/PT  1878  I F (PZ<=0!)  1910  GOTO  18S0  SZ(7,I)=SaR(PZ/PT)  1890  SY(7,1)=2!*ABS(SZ(7, I ) )  : SZ(9,I)=PT  : SX(7,I)=PT : GOTO  1920  SX(9,I)=0!  : SY(9,I)=0!  : SZ(9,I)=0!  SX(7,I)=01  : SY(7,I)=0!  : SZ(7,I)=0!  GOTO  1940  S Z ( 8 , I ) = Q T : SY(8.1)=OD/QT SX(8,I)=0!  1950  NEXT I  1750  : SZ(4,I)=R  SX(6,I)=FNSL(1,PX) : SY(6,i)»Q  1940  1730  E L S E S X ( 3 , 1 ) = F N S L ( 0 , PX1  0 > S Y ( 4 , I ) ) GOTO  SX(4,I)=FNSL(1,PX) : SY(4,I)=0  I F (QT=0!)  1700  : SZ(2,I)»R  : SZ(3,I>=R  I F (PX<XN(1) OR  1840  1930  ";  : VP=A3S<XMA-?X)  S < S Y ( 0 , I 1 ) GOTO  SZ1IM,I)=R  1920  : QZ=0!  1660  S X ( 5 . I ) » F N S L ( 1 , PX) I F (PX>XMV  1820  1910  : OD=01  1640  I F (VP>ZP) GOTO  IM=IM-1  1900  : IM=9  : YP=ZP  ";AN(I);AF;"  ; INPUT  : PD=PD-Q  1810 1830  : PZ=0!  : WP=A3S(PX-XMV)  I F (V(P>YP) GOTO  1790 1800  : SY(6,I)=-10!  "I",#3,"9:"*AN(I)*AF  1590  1640  : LPRINT  AF=".CSP"  1520 FOR  1630  : K2=0  KEY GOTO 1500, 1510, 2560  1530  1620  : I=MQ-1  SX(20,I),SY(20,I),SZC20, I)  1470 LPRINT 1480 ON  : XN( 2) = XN< 2 ) - X N ( 1 ) : B<2>=3(2)•B<1)  : SY(8,I)=0!  : SX(8,I)=3!  : GOTO  1950  : SZ(8,I)=0!  1960 L P R I N T C H H S U 8 ) ; S ? C ( 2 0 ) "Name  : " ;AM;"  [ S U B J E C T DATA  -•;  KEY-2;"]" 1970  LPRINT  : LPRINT S P C ( 1 0 ) " ( " ; K E Y - 1 ; • ) F o r ";CHRS(14);  AF;CHRS(28); 1980  LPRINT  " files  :•;CHRS(27);CHRS(SS) : LPRINT  1990 L P R I N T S P C ( 1 0 ) " T h e C h o r o i d - R e t i n a l  121  Peak";  : J=0  : GOSUB  2590  2800 LPRINT "Mid-Vitreous"; : J=l : GOSUB 2590 2010 LPRINT "Vitreous Minimum"; : J = 2 : G0SU3 2590 2020 IF KEY=0 THEN LPRINT •(Autofluorescencel •; 2030 LPRINT "Lens Peak"; : J=3 : GOSUB 2590 2040 LPRINT "Aqueous Minimum"; : J=4 : GCSU3 2590 2050 LPRINT "Mid-Aqueous"; : J=5 : GOSUB 2590 2060 LPRINT "Corneal Peak"; : J=6 : GOSUB 2590 2070 LPRINT "values c l o s e s t to the 3-, S-, 9-mm , e t c . , a r e 2083 GOSUB 2713 2090 FOR J=10 TO IM 2100 GOSUB 2603 2113 NEXT J 2123 LPRINT CHRS(27);CHRS(58); 2130 IF (KEY=3) GOTO 22S3 2143 LPRINT SPC(13)"The A x i a l Resolution-Ratio are :" 2153 G0SU3 2333 : LPRINT S?C(13)"At 3 mm"; 2163 LPRINT SPC(23)"About 3 mm* ;CHRS (27) ; CHRS< 45 ); CHRSO) 2173 FOB I = XZ TO .10 2180 LPRINT SFC(25); : PD=VAL(AN(I)) : LPRINT USING "»##.#";PD; 2193 PD=SY(11.I)/SY(0,I) : PT-SY(3.I)/SY(3.I) 2233 LPRINT SPC(24); : LPRINT USING "»##. ###";PD; 2213 LPRINT SPC121); : LPRINT USING "##*.*#*";??; 2223 IF KMO THEN LPRINT ELSE LPRINT CHRS (13); 2233 NEXT I 2243 LPRINT S?C(23)CHRS(27>;CHRS(45);CHRS( 1);S?C(72); 2253 LPRINT CHRS(27);CKRS(45);CHRS(3);CHRS(27);CHRS(53) 2263 LPRINT SPC(13)"The Lover Limit of Detection (or S e n s i t i v i t y ) ->" 2273 LPRINT S?C(15)"Average about Mid-Vitreou3 - 2SD"; 2283 LPRINT CHRS(27);CHRS(33);CHRS(1);"ras*;CKRS(27);CHRS(84); ", are :• 2293 GOSUB 2833 : LPRINT "Fluorescein Equivalent (ng/ml)"; 2330 LPRINT SPC(14)"Number•;CHRS(27);CHRS(45);CHRS(3) 2313 FOR I = KZ TO MQ 2323 LPRINT SPC(25); : PT=VAL(AN(I)) : LPRINT USING •#**.#";PT; 2333 LPRINT S?C(25); : ?T=SY (7, I)-SY U , XZ) : LPRINT USING •##•».*»#";?T; 2343 LPRINT SPC(23); : LPRINT USING '#*##.*»#";SX<7,I) ; 2353 IF KMO THEN LPRINT ELSE LPRINT CHRS(13); 2363 NEXT I 2373 LPRINT SPC(23)CHRS(27);CHRS(45);CHRS(1);SPC(75); 2283 LPRINT CHRS(27);CHRS(45);CHRS(31;CHRS(27);CHRS(58) 2393 LPRINT SPC(13)*The R e p r o d u c i b i l i t y Percentages are :" 2433 GOSUB 2833 : LPRINT SPC(13)"Percent";SPC(13)"About"; 2413 LPRINT CHRS127)CHRS(45);CHRS(3) : JJ=3 2423 FOR J=8 TO 9 2433 FOR 1=2 TO HO 2440 IF (SYU, I-1)=0!) GOTO 2520 2450 PD=VAL(AN(I)) : PT=VAL(AN(1-1)) 2463 I F (PD-?T>>4! THEN GOTO 2520 ELSE JJ=JJ-1 2470 LPRINT SPC(21); : LPRINT USING "#.#.#";PT; 2480 LPRINT • -"; : LPRINT USING •###.*•• ;PD; 2493 PD=133!.(SY(J, D - S Y I J , 1-1))/SY (J, 1-1) 2533 2513 2523 2533 2543 2553 2563 2573 2583 2590 2630 2613 2623 2633 2643 2653 2663 2673 2633 2693 2733  LPRINT SPC122); : LPRINT USING •####.##•;?D; LPRINT S?C(9); : LPRINT USING •*##.#";SXIJ,1-1) NEXT I IF (J = 9 AND JJ=3) THEN LPRINT SPC(3S)*No Value can be c a l c u l a t e d ! " NEXT J KEY=KEY-1 : K2=l : GOTO 1473 ERASE XN,3,C,3X,SY, 3Z LPRINT CHRS < 27 ) ; CHRS < 65) ; CHRS (12) ; CHRS (27); CHRS (53) ; CHRS (18) ; CLOSE : ON ERROR GOTO 3 : GOTO 153 LPRINT " values are :" : KW=0 : GOSUB 2710 FOR I = KZ TO MQ LPRINT S?C(25); : PD=VAL(AN(I)) : LPRINT USING "#*#.#";PD; LPRINT SPCI23); : LPRINT USING "####. ###";SY(J,I) ; LPRINT " -/-"; : LPRINT USING "#*».#*••;SZ(J,I); LPRINT S?C(22); : LPRINT USING "#»#.«##";SX(J, I ) ; IF I<>(I0 THEN LPRINT NEXT I LPRINT CHRS (13) ;S?C (23)CHRS(27 ) ;CHRS ( 45 ) ; CHRS (1) ; LPRINT S?C(90)CHRS(27);CHRS(45);CHRS<3> I F J<13 THEN LPRINT CHRS ( 27) ; CHRS (58) ;SPC (13) "The •; RETURN  122  2 7 1 3 G0SU3 2800 2723  QN  KW  : LPRINT  'Concentration  - / - S. £.  2733  LPRINT S P C ( 1 3 ) " D i s t a n c e  2743  LPRINT S?C(13)'Number o f p a i n t s  2753  LPRINT  2763  RETURN  2 7 7 3 LPRINT 2783  (ng/ml)*;  GOTO 2743 from  Retina  (mm)';  : GOTO  2733  entered*;  CHRS(27);CHHS<45);CHHS<0)  "INFO  ERHGR Code  »";ERR;"at L i n e  #";ERL;  I F NESR>13 THEN RESUME 2563 E L S E NERS = N E H R ' l  2 7 9 3 RESUME NEXT 2 8 3 3 LPRINT CHRS ( 1 5 ) ; SPC (23) CHRS ( 2 7 ) ;CHRS ( 4 5 ) ;CHRS ( 1 ) ;  A.10  2813  LPRINT  2823  RETURN  "P.I. Time  (min)';SPC(13);  B L O O D . B A S 1 3 3 3 REM 1013  A="  1323  DIM  Updated  233186  : PRINT  A  : PRINT  *3LC0D"  : PRIST A  : PRINT  A F ( 6 > , I X I 1 3 5 ) . I Y ( 1 3 5 1 , T ( 5 6 ) , Y ( 5 6 ) , W(55), X ( 5 )  1 0 3 0 DIM  ZI5),FR(ll).C(4,a>,HS(4,i3),.-3(4,i3)  1040  DEF  F N C G N C ( X ) = 3 4 . 1 7 « E X ? < - 1 3 9 . 339-SORI11915. 9'X/.114441))  1053  DEF  1863 ON  FNI?U)=CINT(.318315.<L0G<X>'139.339) 2-232.586> A  ERROR GOTO 3523  1373 OPEN  : M=-l  " I * , # 3 , " 3 : S U 3 J Z C T . DAT*  1380 P R I S T T A 3 ( 1 3 ) ; " N a m e 1393 FOR  HP=VAL(AN(K))  1113  I F (HP=3!) GOTO  1123  H=M-1  1133  NEXT K  1143 OPEN  »2.3U  : 3=rNCCNC!GU)  : INPUT  : NU.1=-1 : INPUT  Y(NU«)=rNCGNC<GU)-3  1133  IF iuri2)<>3  THEN C L 3 3 E E L S E GOTO  1233  I F (H?>100!)  GOTO  1213  GU=1!  TO  : INPUT  #2, GV  : W ( SUM)=SSR((FNCCNCIGV))'3-D'2) 1183  : KC = 99  : GOTO  1233  KC  IX(X)=1-CINT(K«GU)  1233  NEXT X TO 3  1270  C(X,S)=0! : C(X,7)=99  1283  NEXT X  1 2 9 0 DEF  *2. GU  : I Y ( 1 3 4 ) = 1 5 : KC=132  1243  K=3  *Z,GU  1223  : IY(134)=i3  K=0  : INPUT  : D=rNCCNC;GV)  H?=T(NU.1)  1133  1222 GU=2!/3!  : HP = 3! #2, GV  #2, T!NUM) : INPUT  I F T(NUM)>K? THEN  1 2 6 3 FOR  : PRINT  1133  " I " , *2, "3:PLASMA.DAT"  INPUT  1233 PGR  : PRINT  : FR(M)=KP  1 1 6 3 NUM = , W « - 1 1173  #3. A F ( 5 )  X « l TG !<C.  1133  1153  : INPUT  : ";AF(5) ; CLOSE  FNFP(J,T)=C(J,3)»C(J,2).T-C(J,4).T*2  1 3 3 3 DEF F N F a < J , T ) = C ( J , l ) . ( T - . 2 S ) * 2 1310  I X ( 1 3 3 ) = C I N T ( F R ( M ) ) : XX=0  1 3 2 8 A F ( J ) = *A ' B«t • C « t * 2 * : GOSUB 1333 FOR  K=3  TO  1343  GU=T(X)  1353  NEXT X  1 3 6 3 GOSUB 2953 X=0  : J=0  : GOSU3  : XX=0  31S3  : IY(3)=0  3113  NUM : GOSUB 3 5 3 0 : E=l!/48!  : XK=KK'l : D=C(J,4)*E-C(J,0)/4I  1373 G V = C ( J , 2 ) / 2 ! : G U = C ( J , 4 ) / 3 ! 1380 FOR  : KY=0  : G0SU3 3 3 3 3  : HP=C(J,3)/4I  TO M  1390  F=FS(K)  1400  FO( J , X)=FNFO( J , F) - H P ' T M - C I J , 5) « < E - F 3 / 3 ! ) 2 - < F N F P ( J , F > ) 2 / 4 !  : HS(J, K)=D-F« ( C ( J , 3)-GV'F-GU'r"2)  1413  GGSU3 3573  A  1423  I F H S ( J , X ) < = 3 ! THEN C ( J , 7 ) = K S ( J , X )  1433  NEXT K  1443 L P R I N T 1450 FOR  A  CHRSI12)  X=i  TO  XC  1463  H?=FNF?(J,K)  1473  NEXT X  1483 GOSUB 2493  : GOSU3 2370  : KX=2 : KY=3  : AFIJM'A  : J=l  - B.log(t)  • C*Clog(t)J*2*  1493 GGSU3 31S3  : GU=C( J , 3 ) - C ( J , 2 )  1500 FOR  NUM  K=0  TO  1510  GU = L0G(T(K>>  1523  NEXT K  1533 GOSUB 2960 1543  A  : GOSUB 3 5 0 0  : E=L0G(2!)  : GOSUB 3 0 3 3  : KK = 0  : GOSUB  3110  :. KK=KK-1  : F=E-2!  : H=F 2 A  : HP=F/4!  : D=H/4!  B=C(J,2).HP-C(J,4).O-C(J,0)/4!  1550 FOR K=0 TO a 1560 E=FR(K) : G U = L O G ( E ) - l ! : H S ( J , K)= 3 * ( F N F ? ( J , G U ) - C ( J . 4 ) ) » E 1570 F0(J,X>=FNFQ(J,Z).C(J,3>«(H?-E.GU>'2-CtJ,5>.(E.<GU 2'l!>-D)'2 A  123  1530 GU=GU-1! : FQ(J,X)= FQ(J,X)•(FNF?(J, GU))A2/4! : GGSU3 3570 15S0 IF HS.'J,K)<=0! THEN C< J, 7) = nSi J, X) 1633 NEXT X 1613 LPRINT CHRSQ2) 1623 FGR X»l TO XC 1630 GU=LCG(K) : KP=FNF?<JG , U) : GGSU3 2373 1543 NEXT X 1650 GDSU3 2453 : KX=3 : KY=1 : J=2 : AF(j)="A-exp(3-t-C»tA2)" 1663 GOSU3 3163 : KK=3 : C(J,3)=EX?(C(J,a)) : C<i,1)»C(J,I)«C(J,3)"2 1673 3=.5'C(J,3> : GGSU3 2333 : GGSU3 311A3 1633 OEF FNFE(i,T)=a-EX?(C(J,2>«T<-C(J,4)-T2) i6sa FOR :<=a TO sua 1703 KK=XK-1 : DW=2!»FNFZ(J, T(K))-Y(X) : GOSUB 3513 1713 NEXT X 1723 GOSUB 2S63 : Z=FNAFS(J,1!> : H=E/2! : G =FNFE(J,.5) : GU=G-H 1733 3Y=C(J,1)•(Gu/C;J,3))2*C (J,A 3)•v H-G/2! iA2*C(i,5 >• (K-G/4!)A 2 1748 KY = 2 : G=C(J, 1)/C<J, 3)2 : H=2!*C(i,4) 1753 0=(G-C(J,3)-C(J, 5)-(C<J,2)-H)A2)'EA2 1763 FOR X=3 TO li 1773 L=CINT(FR(X) : HS(J,K)»GU : FQ(,JX , )=GV 1783 FOR N=KY TO L 1793 F=rNFE(J,N> : HS(J,X)= KS(J,A X)-E-F : HP = NAA2 A 1833 C0=<G-C(J,3>«KP-C!J, 5)«HP2-(C ( J. 2)»H»N)2/4!)-F2 1313 FQ(J,X)=FQ(J,X)-D»CQ : D=CO : E=F 1323 NEXT N 1333 GOSUB 3573 : KY=L-1 : GU=HS(JK , ) : GV=FQ(J,X> 1843 IF GU<=8! THEN C(J,7)=GU 1853 NEXT X 1863 LPRINT CHRSI12) 1873 FOR K=l TO KC 1883 HP=2!"FXF2(J,X> : GOSUS 2373 18S3 NEXT X 1S33 G0SU3 24S0 : XX=1 : KY=0 : J=3 : Ar(J)="A * 3/t • C/tA2" : GGSu'3' 3163 1913 a'u'=CIJ, 3)-2! »Ci J, 2)-3!«C( J, 4) : KX = 3 : GG3U3 3333 : GOSu'3' 3113 1923 FOR X=0 TO NUH 1933 KK=KX-1 : GU=1!/T(X) : GGSU3 3530 1943 NEXT X 1953 GGSU3 2963 : 3=LCG(21>-.5 : H?=CtJ, 2)"3-C(J,3)/4!-3!»C(J,4) : PRINT 1963 FOR X=3 TO M 1973 E=FR(X) : GU=LOG(E) : GV=1!/E : HS(J,X)=HP-C(J,3>"E-C(J, 2)-GU-C(J,4)«GV A 1983 FO(J,X)=FNFO<J,£)-C(J,3)•(GU-3)A2-CtJ, 5)•<31-GV)2 1993 FQ(J,X>=Fa<J,X)-.25"<FNF?U,GV>)A2 : GCSU3 3573 2303 IF HS(J,K><=3! THEN C( J,7)=KS(J, X) 2013 NEXT X 2323 LPRINT CHRS(12) 2033 FOR K = l TO KC 2043 GU=1!/K : HPF ' NFP(J,GU) : G0SU3 2373 2353 NEXT K 2363 GOSUB 2493 : GU=lD-22 :A LPRINT CHRS(IS) : PRINT : PRINT 2373 PRINT "The Reduced CHI2 of the FITs are;" : PRINT 23S3 FOR J=3 TO 3 23S3 PRINT : PRINT " ( ";J-l;"> \ C(J,S) ; " ";SGN(C(J,7)> 2138 IF (C(J,7)<=3! OR C(J,S)»>GU) GOTO 2123 2113 GU=C(J,S> : K=J 2123 NEXT J 2123 PRINT : PRINT "The 3EST fit va3 " : PRINT : PRINT " f = ";AF(X) : PRINT 2143 LPRINT SPC (23) N "AME : ";AF<51;" (Plasma Integration)" : LPRINT 2153 LPRINT : LPRINT : LPRINT S?C(131CHRS(27);CHRS(45);CHRS(1); CHRS(14); 2163 LPRINT "PLASMA FLUORESCEIN •;CHRS(23);"Results"; CHRS(27);CHSS(45);CHSS(9) 2170 LPRINT : LPRINT : LPRINT SPC(23)CHSS(27);CHRS(45);CHRS(i ) ; "Time (min)"; 2183 LPRINT 3?C(13)"Concentration (ng/nl)";CKRS(27!;CnRs(45);CHRS(3) 21S3 LPRINT CHRSI27);CHRS(S3); : XP'li 2233 FOR J=0 TO NUM 2213 DW=T(J1 ; 3W=Y(J) : CW=W)J( : LPRINT SPCI27); : GCSU3 2423 : LPRINT 2223 NEXT J 2233 LPRINT : LPRINT : LPRINT SPC(13)CHRS(14);"The BEST fit was <•>" 2243 LPRINT : LPRINT S?C(12)CHRS(14);"f = ";AF(K);"." 2253 OPEN "0",#3, "3:?LASflA. FIT" : PRINT #3, X 2263 F OR J=0 TO 5 STEP 2 2273 PRINT #3,CCK,j> : GU=SCR(A3S1CI K, J'l))) : PRINT #3G ,U 2283 N EXT J 124  2298 FOR J=0 TO N 2380 PRIHT *3,FR(J) : PRINT #3,HS(K,J) : PRINT #3,FO(K,J) 2310 NEXT J 2320 ERASE AF. IX. IY.T.X, Y, Z. H.FR.C. HS, FQ 2330 ON ERROR GOTO 3 : LPRINT CHRS(27);CHRS(53); 2340 CLOSE 2358 KILL "3P : LASMAD . AT' 2360 GOTO 150 2370 IF HP>108! THEN GOTO 2400 ELSE IY(K>=3 2380 IF (K<IX(133) AND HP<0!) THEN C(J, 71-HP 2390 RETURN 2400 IY(K)=FNI?(H?) 2413 RETURN 2423 LPRINT USING •##*.#•;DW; 2430 LPRINT SPCIK?); 2453 LPRINT • •/-'; 2463 LPRINT USING '#». 2473 RETURN 2433 REN Subroutine for printer-plotting. 24S3 G0SU3 2333 : LPRINT CHRSI15>C ; HRS<27>C ; KRS(49); 2533 LPRINT SPC123) 'Vertical (LOG) Scaie = X 13 ng/isl / 13 div. "; 2513 LPRINT S?C( 13)'Horizontal Scaie • ' ; IY( 134 ) ;'min / 13 div. • 253 23 FO KY R=K5=30-KTO: K5P3= XY/5! : N=8 : L--1 : LPRINT CHRS(13); CHRSU3) ;CHRS(9); 2543 IF ((HP-FIX(HP))=3) GOTO 2563 2553 LPRINT CNRSI9);CHRS(124); : GOTO 2638 2563 IF (K=1S OR K-25 QR X»35 OR K = 45 OR X = 5) GOTO 2593 2573 IF K=3 THEN LPRINT 'ng/ml -'; ELSE LPRINT CHRSI9); • »•; 2533 GOTO 2533 2593 LPRINT ' 13;'CHRS(27);CHRS<83>;CHRS(3);2-XYN13; CHRSI27);CHRS(34);' -"; 2633 FOR KZ=3 TG NUN 2613 IF FNI?(Y(KZ)=KY "KEN L=KZ 2623 NEXT K2 2633 IF !L=-i) GOTO 2713 2643 FOR XZ=3 TO NUN 2553 IF FNI?(Y(KZ) )<>KY THEN GOTO 2S93 ELSE XX=1-CINT(7(KZ) 2663 IF XX=9 THEN XX=1 ELSE XX=KX-N 2673 IF XX<>3 THEN LPRINT SPC ! XX-i ) "a"-; 2633 N=CINT(T(XZ)-l 2653 NEXT XZ 2733 N=9 : LPRINT CHRSI13);CHRS(9);CHRS(9) ; 2718 FOR XZ=3 TO KC 2723 IF IY1XZJOKY THEN GOTO 2763 ELSE XX=IX(KZ) 2739 IF XX=3 THEN XX=1 ELSE KX=KX-N 2743 IF XX<>8 THEN LPRINT SPC(XX-1)CHRS(249); 2753 N»IX(XZ) 2763 NEXT XZ 2773 NEXT X 2783 LPRINT CHRSI 10) C ;HRS< 13) ;CHRS(9) ;CHRSi9); • •; 2793 FOR K=8 TO 9 2803 LPRINT s '•'; 2813 FOR N0 TO 8 2820 LPRINT 2833 NEXT X 2843 NEXT X 2853 LPRINT ••• : XY-2-IYI134) . LPRINT CHRS I 9) C ; KRS (9) ; 8; 2863 FOR X«l TO 4 2873 LPRINT SPCI16)KY-X; 2883 NEXT X 2893 LPRINT S?C( 17)'aiin" C ; HRS (27) C ; KR3 < 58) ; CHRS127);CHRS(S5);CHRSI12); 2930 LPRINT CHRS 127); CHRS (58) t LPRINT : LPRINT SPC (41) "CHI "2 = ' 2913 CW=C(J6 ,) : GOSUB 2463 : LPRINT 2920 IF XK>3 THEN CIJ, 6)-C(J,6)/(KX-3) 2933 LPRINT SPC (33) R ' educed CHI "2 - "; : CW=C(J,S) 2940 GOSUB 2460 : LPRINT CHRSI12) 2950 RETURN 2963 LPRINT CHRSI 13) : LPRINT : LPRINT SPC(10)CHRS(27); CHRSI45);CHRS(1); 2973 LPRINT R ' ESULTS of INTEGRATIONC ;' HRS127)C ; HRSI 45);CHRS(0) 2980 LPRINT CHRS(27);CKRS(S8) : LPRINT SPCI151CHRSI27); CHRS145);CKRS(1); 2990 LPRINT "For 3 < t < T »in. AREA •/- S. E. Imin-ng/ml) *; • 3330 LPRINT SPCI6) 'Percent (X)";CKRS(27);CKRs<45);CHRS(3) 3318 RETURN 3823 REN Subroutine for printing curve-fitting coefficients. 3833 LPRINT CHRSI 18);CHRS(27);CHRs(65);CHRSI12);CKRS(27);CHRS(S3) H*tt'---';Cil:  125  3040 LPRINT SPCIZaiVMAHE : •;AF(5);' (Plasma Integration)' : LPRINT 3053 LPRINT : LPRINT SPC(10)•(•;SIGHTS(STRS(J-l),1); •) Fitting to i = *;AF(J) 3363 LPRINT SPC(25)"A = "; : 3W=C(J,3) : CW=SCS(A3S(C(J,1))) : GGSUB 2440 3070 LPRINT : LPRINT S?C(25)'3 = •; : 3W=C(J,2) : CW=SCH(A3S(C(J,3))) 33S3 3CSU3 2440 : LPRINT : LPRINT 3?C(25)"C = : 3W=C(J,4) 3093 CW=SCniA3SlC<J,5)) : GGSU3 2443 : LPRINT : LPRINT : LPRINT 3133 RETURN 3113 LPRINT CKRS(27);CHRS(53) : LPRINT SPC(23)CH3S(27); CHRS(45);CHRS(1); 3123 LPRINT "Time (min) Data (ng/ml) CFit-Datai Percent IX)'; 3133 LPRINT CHRS(27);CHXS(45);CKRS(3>;CHRS<15> 3143 RETURN 3153 REM Subroutine for curve-fitting to oroer 2. 3163 FGR X=0 TO 4 3173 X(X)=3! : ZC<)=3! 3180 NEXT X 3193 FOR X=0 TQ NUM 3233 ON KX GOTO 3220.3230 3210 HP=T(K) : GOTO 3240 3223 HP=1!/T(K) : GOTO 3243 3223 HP=LOG(T(K>) 3240 ON KY GOTO 32S3 3253 ' GV=Y(X) : GU=I!/W(X)*2 : GOTO 3280 3263 IF (Y(X)<=3!) GOTO 3353 3273 GV*LCG(Y(K) : GU= ( Y( K > W / ( K) )A2 3280 Z(4)=GU : Z(3)*GV.GU 3293 FOR N=3 TO 4 3333 X(N)*X(H)-Z<4> : ZI4)=Z(4)»HP 3313 NEXT N 3323 FOR 1 .=3 TO 2 3333 Z(N)=Z(N)-Z(3) : Z(3)=Z(3)'HP 3343 NEXT N 3353 NEXT X 3363 HP=X<1)/X(3> : 3U=X(2)/XI1) : GV=X(3)/X(2) : 3=X(4)/X(3) : G=GU-G7 3373 D = l!/GV-i!/GU : E=l!/HP-i!/GU : ZI 3)=G-D-(3-GV)•£ 3333 F»Z(0)/X<1)-Z(1)/X<2) : C(J, 4) = <F-D-(Z(2)/X(3)-Z(1)/X(2))*E)/Z(3) 3333 CIJ,3)=IF-C(J,4)'G)/E : C(J, 2>=Z(3)/X11)-C(J, 3)/HP-C(J,4>«GU : F=3-GJ' 3400 H=(B»(GU-HP) /GU'HP-GV-G»HP/ GV)"2 : C0=3-GV A 3413 C(J. l)=C0A2/X(3)-2!•HP 2»C0'F/X(2) A 3423 CIJ, i)=C(J, l)-(HP'F)2/X(2)-2!'HP-G-CO/(X(3)-GV)•(HP'G)"2'3/X(3) 3433 C(J,l) = (C(J.l)-2!>H?A2«r"G/X(2))/H : CO = X(4)/X(2)-XI 2)/X(3) : G=GV-HA? A A 3440 CIJ,3)=(F/GU>2/X(0)-CO2/(X(3)'GV)-3«G2/X(3)-2!F-C0«HP/(GU«X(3>) 3450 C(J,3)=<C(J,3)-21'G-F'K?/X(3)-2!-Ca'G/X(3))/H : F=GV-GU : CO=GU-HP 3460 C!J,5)=(FAX(1)/X(3))A2/X(3)'G"2/(X(3)'GAV)'3.(C0/GU)A2/X(3) 3473 C(J,5)=(C(J,5)-2!'(G-C0)-F'H?«X(1)/X(3)2-2!>G«C0/(GU«X(3)))/H 3483 RETURN 3433 REM Subroutine for printing deviations. 3533 SW=FNF?(J.GU>-Y(X) 3513 LPRINT 3?C(33); : LPRINT USING •#«#.##•;T(K); : LPRINT 3?C(8); 3523 CW=Y(X) : GGSU3 2463 : LPRINT S?C(3); : C W = DV : GGSUB 2463 3533 LPRINT S?C(7>; : CW«100«!DW/Y(X) : GOSUB 2463 3543 LPRINT : C(J.5)=CiJ,S)-iDW/W(X))"2 3550 RETURN 3563 REM Subroutine for printing resuit of integration of fit. 3573 ?0(J,X)=SSR(A3S(rB(J,X))) : LPRINT S?C(18)"T = " ; : XP=8 : DW=F3(K) 3583 3«-KS<J,X> : CM=FSiJX r J : G GSuo 2423 : LPRINT CHRSI15);" => •; 3590 CW=130!«FB(JK)/KS(J,K) : G0SU3 2463 : LPRINT CHRS(27);CKSSI5S) 3633 RETURN 3613 REM Error subroutine 3623 IF NERR>13 THEN END ELSE NE3R=NERR-1 3633 LPRINT "3LCQD ERROR Code #";ERR;"at Line #";ERL; 3643 RESUME NEXT f  126  1 1  C/VAZ.BAS  1000 REM Updated 123386 1818 DI!t A£(2),H(2),R<2S6),S(25S),U(256>,T(10>,a(6>,ZH(2) 1323 DM I X(2,256),Y<2,256),Z(2,255),W(2),TG(2),0(2),P(2),US(3> 1833 A= ••***•• : PRINT 1848 QN ERROR GOTO 3P4R 6IN 3T A : PRINT 'C/VAZ'PRINT A 1358 OPEN 'I',#3,•BiSUBJECT. DATIN ' PUT #3, A W : PRINT TAB(18)N ;' ame (A •;W , : INPUT #3A , INPU : T #3, A 1363 INPUT #3A 1873 INPUT #3,0(8) : INPUT #3,0(1) : INPUT #3,0(2) PUT #3,US(3) : INPUT #3,USUI : INPUT #3,US(2) 1888 IIN NPUT #3, W W : O(5)=250«!WW : SS=US (0)/Q( 0) 1 8 9 8 1138 CLOSE : NC=1 : Q( 3)3 (US( 3) *US (1) *US(2)) / 2! 1113 KP=0 : NC=NC-1 : J=-l : KC=3 1123 AE(3)=AN(1) : OO=VAL(AE0!>>AE :tl)=AN(NC) 1138 OPEN 'I",#2,"3:PLASMA. FIT": INP UT #2,1 ?P = VAL(AEU) ) 1=3 TO 5 1148 FOR INPUT »2,CV 1158 N 1163 EXT I , Y : INPUT #2C ,W : INPUT #2, CV 1173 INPUT #2Y 1183 IF (YY=aO OR YY=PP) THEN = J-1 EALSE GOTO (J)=63!.CW : CV=60-!CV : P(J)=CV2 1193 0 IF J=l THEN CLOSE ELSE GOTO 11731173 1 2 0 0 1218 LPRINT CHRS(27);CHRS(65);CHRS(12);CHRS<27>;CKRS(S31; 1223 GCSU3 1913 : ?P=0(3) : PRINT 1233 LPRINT SPC(8)"Centre of Retinal Curvature"; 1243 GOSUB 1953 : LPRINT : PRINT ' LU0RESCEIN Used =';0(5);' mg ." 1253 LPRINT SPC(19)F 1263 LPRINT : LPRINT : LPRINT 1273 LPRINT SPC(5)CHRS(14);'!";RIGHTS(STRS<KC'1),1); •) ALG INMENT by "; KC=3 THEN LPRINT "RETINA" ELSE LPRINT "C/R PEAK" 1283 IF 1293 IF XC=3 THEN AP=".RE7" ELSE AP=".CRP" 1338 OR 1=3 TO 1 1313 LPRINT : LPRINT : A="3:"*AE(I)*AP : OPEN "I",#1,A 1328 L=-l : TT=3! : RR=1E*22 : XX=RR : PRINT TA3I13); •Reading ';A : PRINT 1338 L=L-1 : INPUT #1,R(L) : X(I,L)=R(L) : INPUT #1,S(L) 1348 Y(I,L)=S(L) : INPUT #1, YY : U(LXYYA2 : Z(I,L)=U(L) 1353 00=SS-R(L) : QQ=ABS(00-3)! : PP=ABS100) 1363 IF Q ( O>RR) GOTO 1383 1373 RR=00 : WW=SL () 1383 IF (PP>XX) GOTO 1400 1398 =PP : W(I)«S(L) : TG(I)=U(L) 1433 IFXX(0 0>31 OR S(LXTT) GOTO 1420 1418 T T = S(L) : ZH(I)'RIL) 1423 IF (EO (1)=0) GOTO 1333 1438 CLOSE : LF P INT CHRS(18) : LPRINT SPC(20)'File : •; 1443 LPRINT CHR R IXAP;CHRS<20>;" has";L-1;"data-seta. 1453 LPRINT : N(I)S=(1L4);A:E(K EY=0 : GOSUB 2B10 1468 NEXT I 1470 00=W3(W /) W : LPRINT CHRSI18) : LPRINT SPC(5)CHRS<14); 1480 LPRINT 'C/R CORRECTO I N CONDITION";CHRS(28);• *"; 1498 GOSUB 2020 : PP=11 : OQ=0! : WW=1! :A GOTO 1573 1533 WW=W1(W /) <3) : XX=TG(1)/W(1)"2-TG<3)/W(3)2 : TT=3! 1S13 FOR X=3 TO NO) , )=YY"WW 1523 YY=Y(8,X) : Y(0X 1533 IF (SS»X<3,X)>3! OR Y(8,XXTT) GOTO 1553 1543 TT=Y(8,X) : ZH(0)=X(3,X) 1553 NEXT K 1563 SO=WWS.CRX (X) : P=WW 1573 GOSUB 1913 : LPRINT SPC!6)CHRSI14) ; 15ae LPRINT "For a C/R RATIO"C ; HRS<23); • ="; 1590 G0SU3 2300 : LPRINT CHRS(18) 1633 IF KC=0 THEN GOSUB 2100 ELSE GOSUB 2050 1613 LPRINT S?C(10)CHRS(14);AE(l);CHRS(20);"-minute minus •; 1620 LPRINT CHRS(14);AE(3) ;CHRS(28);"-minute scan leaves'; 1630 LPRINT L'l;"data-sets." : LPRINT 1648 1=1 : GOSUB 2210 : J5=0 1650 FOR K=0 TO L 1663 IF (R<K)>0(3)> GOTO 16S0 1670 IF (R(K)>0! AND S(X)>0!) THEN J5=J5-1 1680 NEXT K 1690 IF (J5>11) GOTO 1720 1700 LPRINT CHRS(18);SPC(16)"0nly';J5; •data-sets => no file was created.• 1710 GOTO 1840 J  127  1723 IF WW=1! THEN K = 3 ELSE X»l 1733 KEY=2*KC-1 : AA= *.CV*-RIGHTS(STRS<KK ' EY>,1) 1743 A="3:"*AE(I)-AA 1753 OPEN '0\»3,A 1763 FOR K = 8 TO L 1773 IF (R<K>>8<3>> GOTO 1813 1783 IF (RIKX81 OR S(XX=8! OR U(KX=8I) GOTO 1833 1793 OO-SBRIUIK) : PRINT #3,R!K) : PRINT #3,S(K> : PRINT #3,00 1833 NEXT X 1813 CLOSE : LPRINT CHRSU8) ;S?C(13)J5; 'data-sets »ere saved in •; 1823 LPRINT CHRSI14)A ; : LPRINT : LPRINT : LPRINT 1833 LPRINT SPCI25)'Deleted : Yes [ 1 No C J* 1843 IF Ww>!l THEN GOTO 1533 ELSE KC=KC-1 1853 IF (KC=1) GOTO 1213 I860 IF <NC<MO) GOTO 1119 1873 ERASE AE, N, R, S, U, T, 3, ZH, X, Y, Z, W, TG, 0, P, US 1883 LPRINT CHRS (27); CHRS 158); CHRS (27); CHRS (65); CHRS (12); 1893 ON ERROR GOTO 3 : LPRINT CHRS( 27) ; CHRS( 53) 1930 CLOSE : GOTO 153 1913 XP=KP-1 1923 LPRINT CHRS (12); CHRS( 18) ;SPC ( 25) N "AME : *;AW; • I C/VAZ -';KP;']" 1933 LPRINT : LPRINT 1940 RETURN 1958 LPRINT • •;CHRS(247); • •; 1960 LPRINT USING •##.•#####• ;PP; 1970 LPRINT • na f ran. the RETINA. ' 1980 RETURN 1993 LPRINT SPC(40); : LPRINT USING "##.###'"•;00; : LPRINT SPC120); 2000 IF PP=-2.2E-22 THEN G0SU3 3360 ELSE LPRINT USING 2013 LPRINT '•/-'; 2023 IF QQ--2.2E-22A THEN GOSUB 3360 ELSE LPRINT USING "»#.#####""";C0 2030 RETURN 2043 REN Subroutine ta (re-order and) subtract data-sets. 2050 FOR J*3 TO 1 2363 FOR X=0 TO N(J) 2373 X( J,XXX(J,X)-ZH!J> 2383 NEXT X 2393 NEXT J 2133 L=-l : «=8 2110 FOR K=0 TO N(l) 2123 IF (X(1,X)>X(0,N(0))) GOTO 2190 2130 FOR =iM TO N(0) 2140 IF X( 1, K) <>X(0, J) THEN GOTO 2170 ELSE L=L-1 2150 S(L)=Ytl,X)-Y(3, J) ; R(L)=X(1,K) : U(L)= Z(1, X)-Z(3,J) 2163 M=J-1 : GOTO 2183 2173 NEXT J 2183 NEXT X 2198 RETURN 2238 REM Subroutine for printer-plat. 2218 LPRINT CHRSI15) : LPRINT CHRS(27);CHRS<49); 2223 LPRINT SPC125)'Vertical (LOG) Scale * X 13 / 10 div.'; 2233 LPRINT SPC (131'Horizontal Scale > 1 u / 11 div.'; 2240 PJ=0t : RR*Q(0XO<1) : Q0=RR-a(2) 2250 FOR K=3 TO L 2263 IF <R(K>>0(3>> GOTO 2283 2273 R(X)=SS«R(X) : YY=R(X) : GOTO 2333 2280 IF (R(X)>RR) GOTO 2333 2298 R(K)»US(l)»(R(K)-a(3))/a(l)-YY : 00=R(K) : GOTO 2343 2333 IF (R(K)>flO) GOTO 2323 2313 R<K)=US(2)MR(K)-RR)/S(2)*00 : TT=R(X) : GOTO 2348 2323 R(K)=PSCALE»<R(K)-OQ)-TT 2333 IF <R(K)<0(3) AND S(K)>PJ) THEN PJ=S(K) 2343 NEXT X 2358 JY=3 2353 IF PJ>.8364 THEN JY=CINT(. 018315*(LOG(PJ)-109.339)*2-212.536) 2370 FOR K=0 TO 30 2388 J=33-K : YY»J/5! : M=3 : LPRINT CHRS(13);CHRS(13);CHRS(9); 2393. IF ((YY-FIX(YY))=0!) GOTO 2410 2433 LPRINT CHRS(9) ;CHRS(124); : GOTO 2463 2410 IF (X=5 OR K=15 OR X=25) GOTO 2448 2423 IF K*8 THEN LPRINT 'ng/inl •"; ELSE LPRINT CHRS< 9); * •" ; 2438 GOTO 2463 2443 LPRINT " 18*;CHRS(27);CHRS(83);CHRS(8);INT(J/13); 2453 LPRINT CHRS(27) ;CHRS(84); * <••; 2463 IF (J>JY) GOTO 2573 •##. t t t t l " " " ; ? ? ;  128  2473 FOR J5=3 TO L 2483 XW=0 2493 IF S(JS)>.3364 THEN KH-CINTl.313315' (LOG 1S(J5))'139.339)'2-212. 536) 2533 IF XWoJ THEN GOTO 2563 ELSE XL-CINTI13! «R( J5))-1 2513 IF (KL>U3 OR KL<0> GOTO 2563 2523 IF KL=0 THEN XR=1 ELSE XR=KL-H 2533 IF <KR=0> GOTO 2553 2543 LPRINT S?C(KR-1)CHRS<249>; 2553 M=KL 2563 NEXT J5 2573 NEXT X 2533 00=lE-22 : RR=3! : 00=00 : TT=RR : PP=111 : YY=0! : LPRINT 2593 FOR K=3 TO L 2633 IF !R(K)>PP) GOTO 2673 2613 'IF (R(KXYY) GOTO 2663 2623 IF S(X)>RR THEN SR = SIK) 2633 IF S(X)<00 THEN 00=S(X) 2643 IF R(K)>TT THEN TT=R(X) 2653 IF R(K)<00 THEN 00=R(K) 2663 NEXT X 2673 PP=3! : LPRINT CHRS(9);CHRS(9); • "; 2633 FOR K=3 TO 113 2693 IF (K/IBOPP) GOTO 2713 2733 • LPRINT : PP=PP*1! : GOTO 2723 2713 LPRINT 2723 NEXT X 2733 LPRINT CHRS(9);CHRS(9);" R";S?C(19)"2*;SPC(19)"4"; 2743 LPRINT SPC(19) "6" ;S?C( 19) "8" ;SPC(13)113';SPCI8) -M' : XEY=1 2753 LPRINT : PP=OB : LPRINT SPC(51) "LEFTMOST point*; : GOSUB 1953 2763 LPRINT : PP=TT : LPRINT SPC(53)"RIGHTMOST point'; : GOSUB 1953 2773 LPRINT : PP=RR : LPRINT 5?C(44)"MAXIMUM"; : GOSUB 3423 2783 LPRINT : PP = Q0 : LPRINT SPC(44)"MINIMUM" ; 2793 GOSUB 3423 : LPRINT : LPRINT : LPRINT 2833 LPRINT CHRSI27);CHRS(65);CHRS(12);CHRS(27);CHRS(S3);CHHS(13) 2813 XW=0 : XR=8 : TT=lE-22 : GOSUB 3333 2823 XX=-99 : XO=KK : J5=KX : KT=KK : KL=XT 2833 FOR K=0 TO L 2843 IF XEY=1 THEN 00=R(X) ELSE 00=SS"R(K) 2853 IF 00>6.5 THEN GOTO 3313 ELSE C0=ABS(31-00) 2863 IF (00>TT) GOTO 2883 2878 TT=00 : X T=K 2888 IF (S(KX=8!) GOTO 3388 2393 IF 00<2. 5 THEN GOTO 2978 ELSE KW = KW»1 2933 IF XW=1 THEN J5=K ELSE XL=X 2913 PP=(B(3)-00)*2 : T(3)=S(X)»PP : XX=1.59293E-33.00A2-. 3381543 2923 PP=41-(0(4)-XX)/PP : T(5)=(U(K)/S(K)A2-PP)*T(3)*2 2933 IF X (W = 1) GOTO 2963 2943 PP=T(2XT(3) : Q0=00-T I4) : RR=PP"0Q/2! : T(8)=T(8)*RR 2953 Tll>"T(l)-((T(5)-T(S))/PPA2-(T(9)-XX)/QQA2)«RRA2 2963 TI4X00 : T(6)»T(5) : T(2)=T(3) : T(9)=XX 2978 IF (00<21 OR 00>4!) GOTO 3888 2988 T(7)=T(7XS(K) : T(8)=T(8)-U(K) : KR=KR-1 2998 IF KR=1 THEN XK=K ELSE KQ=K 3338 NEXT K 3313 FOR X=2 TO 6 3323 T(X) = -9.9 3833 NEXT X 3348 IF XEY=1 THEN PP=1! ELSE ?P=SS 3350 IF J5*>3 THEN T(2) =R(J51 "PP 3863 IF KL=>3 THEN T(3)=R(XL)«PP 3873 IF XK=>8 THEN T( 4S) =R (XK) "PP 3383 IF KO=>0 THEN T(5)R(XO)*PP 3393 IF XT=>3 THEN T(6)=R(XT).PP 3188 LPRINT SPC( 18) CHRS (27) ; CHRS (43) C ; HRS( 1) ; "PERMEABILITY "; 3113 LPRINT "COEFFICIENT t PENETRATION RATIO"; 3123 LPRINT CHRS(27);CHRSA(45) C ; HRS( 3);CHRS(27);CHRS(53) 3133 LPRINT ; YY=P(I)/0(I)2 : LPRINT SPC(15)"Mean PERMEABILITY •; 3140 LPRINT "COEFFICIENT (cn"; : J5=l : GGSU3 3308A : SS=-2.2E*22 3158 LPRINT SPC(33)"is •; ; PP=. 32»T(3)/(0(I)-0(3) 2) 3163 IF (T(3X>0!) THEN OB=PP»SBR ( Tt1)/T<8>A2«YY»4!"B<4>/0(3)A2) 3178 GOSUB 2300 : LPRINT : LPRINT SPCIlSJ'Mean PENETRATION RATIO (•;  129  3180 G0SU3 3330 : LPRINT S?C(33)"i3 3193 IF <KR=3) GOTO 3213 3230 PP-Tt7)/(Q(I).KR) : C5=?PS .CR (T<8)/T( 7) *2-YY) : GOTO 3223 3213 PP = -2.2£-Z2 : S==PP 3223 GC3U3 2300 : LPRINT 3233 LPRINT S?C( 15) "PENETRATION RATIO </s), defined at"; 3240 IF (KT<0> GOTO 3270 3230 J5=6 : GCSU3 3330 : PP=S(KT )/0( I) : LPRINT S?C(30)*i3 32S3 aB.=A3S<PP>.SSR(YY.U(KT)/S<XT2),') : GGSU3 2300 : GOTO 3230 3270 GOSUB 3360 : GOSUB 3340 : GOSUB 3363 : LPRINT *•/-"; : G0SU3 3360 3230 LPRINT : LPRINT : LPRINT : LPRINT 3290 RETURN 3300 LPRINT "/a), defined from"; : J5=J5»1 3313 IF T1JSX3! THEN GGSUB 3360 ELSE LPRINT T(J5); 3320 LPRINT "to"; : J5=JS-1 3330 IF TCJ5X01 THEN GGSUB 3360 ELSE LPRINT T(J5); 3343 LPRINT "mm , ' 3350 RETURN 3363 LPRINT " ••. ; 3370 RETURN 3383 FOR X=3 TO 9 3390 TIKX0! 3400 NEXT X 3413 RETURN 3423 LPRINT " concentration *;CK3S<247>;* •; 3433 LPRINT USING "### .#### ;?P; : LPRINT " ng/ml ." 3443 RETURN 3453 REM Subroutine for error printing. 3463 IF (ESHool) GOTO 3493 3473 CLOSE : A=" A:"-AE( I)-AA 3483 RESUME 1753 3490 IF NERR>13 THEN RESUME 1873 ELSE NERR = NERR*1 3533 LPRINT "C/VAZ ER3GS Cade #";£3R;"in Line #";ERL 3513 RESUME NEXT •;  •;  A. 1 2 SLOPES.  BAS  1000 REM Updated 090386 1010 DM I T(5),*<5),3(4),E(4),P<13>,S<13>,U<13),C(13,3), 0(13.3).0(13.4).H(13.3) 1023 DM I X(230),Y(200>,Z!200>,3(13,9),8(13.9).G(13,S). F(13.S),H(13,3),V(4,6) 1030 A-*..".." : PRINT : PRINT A : PRINT "SLOPES* : PRINT A : PRINT 1040 ON ERROR GOTO 2310 : NERR0. 1050 OPEN *I\#3."3:SU3JECT.DAT" : INPUT #3, AE 1060 PRINT TABUS); N ' ame : ";AE : PRINT 1070 INPUT #3, A : INPUT #3A , : INPUT #3A . 1080 INPUT #3S ,O : INPUT #3.SS : INPUT #3,?P 1090 INPUT #3,?P : INPUT #3,3S 1100 CLOSE : B(3)=RR/3!»PP : B(0XPP/3! : SC=PP/CO 1110 3<1)*2!«B(0) : BI2XPP-31 : SL=RR/SS 1120 FOR 1=9 TO MQ 1130 P(I)*VAL(AH(I>) 1140 NEXT I 1130 OPEN "I", #2.'3:PLASHA.FIT* : INPUT #2,1 : J»l 1160 FOR 1=0 TO 5 1170 INPUT #2. VV 1130 NEXT I 1190 INPUT #2V , V : INPUT #2. SS : INPUT #2. PP 1230 IF (VV<>P(J)) GOTO 1223 1210 SIJX631.SS : U(J)"63!«PP : J-J'l 1223 IF E0F(2X>9 THEN CLOSE ELSE GOTO 1190 1230 FOR 1=9 TO HQ 1240 A«'3:".AN(I).*.AVG* : OPEN "I*,#2,A : PRINT 1259 J—1 : BB'0I i CC=0t : SS-91 : SG-3t»BC3> : SD-SG/SC 1269 IF (EOF(2X>0 OR JJ«199 OS SS">B<3)> GOTO 1350 130  INPUT #2, PP : INPUT #2,00 : INPUT #2, RR IF PP>SD THEN SS=SG»(PP-SD>ASL ELSE SS-SC-PP IF <SS<"91) GOTO 1260 JJ-JJ-1 : X(JJ)=SS : YUJ)=QO : Z(JJ)=RR 1310 IF (SS>B<0) OR QQ<BB) GOTO 1330 1320 E(0)=SS : BBS 'Q 1330 IF (SS<B(1) OR 00<CC) GOTO 1260 1340 CC=00 : E(3)=SS : G OTO 1260 1350 CLOSE : PRINT •(•;I-l;•)•;JJ*1;'data-sets ure used from •; 1360 PP=E(3)-£(0> ': E(1)=PP/3I : E(2)>2I«E(1) : 0(I,3)"JJ»1 1370 E(l)*E<a)»E(l) : E<2)=E<0>-El 2) : (1=0 1380 FOR J=0 TO 2 1390 PRINT TAB! 10); ; 1400 IF J=l THEN PRINT •HID-"; 1410 IF J<=1 THEN PRINT P'OSTERIOR '; ELSE PRINT A ' NTERIOR '; 1420 PRINT V ' ITREOUS •••• : VW=1E1.2 : 11=0 1430 FOR XK=0 TO 3 1440 GOSUB 2540 : 22=01 : YY=0! 1450 FOR X=H TO JJ 1460 IF (X(K)>E(J-1)) GOTO 1500 1470 IF X(KXE(J) THEN GOTO 1490 ELSE PP-X(K) 1480 GOSUB 2930 : ZZ=ZZ-1! : YY=YY-((Y(K)-BB)/Z(K))A2 1490 NEXT X 1500 IF ZZ>,21 THEN YY*YY/(ZZ-21) 1510 PRINT Fit';XK;'=> Reduced CHI'2 =';YY 1520 IF Y (Y>WW> GOTO 1540 1530 WV=Y *Y : KC=KK : XX=ZZ 1540 NEXT KX 1550 H=K : 11=1 ; 0(I,J)=XX : H(I,J)-*K : XK=KC : N(I,J)=KX 1560 FOR K=0 TO 2 1570 NN=3«J«K : L=2"X : 011, NN)=V(KCL , ) : R(I, NN)=V(XC,L-l) 1580 NEXT K 1590 PRINT TAB(10);'9est for fit =';KC : PRINT 1600 IF J = 2 THEN PP=B(2), ELSE PP=6l«J<-31 1610 GOSUB 2930 : C(I.J)=BB : D(I,J) = DD : G(I,J)=GG 1620 F(I,J)=HH : l'J-3 : GU.L)=FF : F(I,L)=EE 1630 NEXT J 1640 NEXT I 1650 GOSUB 2240 : LPRINT SPCI25)CHRSI14)R '; EDUCED CHIA2' 1660 LPRINT CHRSI15) : GOSUB 2300 1670 FOR 1=0 TO HO 1680 PP=P(I) : GOSUB 2430 : LPRINT • •; 1690 FOR J=0 TO 2 1700 00=H(I,J) : GOSUB 2490 : LPRINT • -';INT!0<I,J)); •-';N(I,J);SPC(4); 1710 NEXT J 1720 LPRINT ' ';INT(0(I,3)) 1730 NEXT I 1740 LPRINT CHRSC18) : LPRINT SPC (25) CHRSI 14)"CONCENTRATION;' 17S0 LPRINT CHRS(20);" - C"; : A-'-l" 1760 GOSUB 2270 : LPRINT " 1" : 11=0 : GOSUB 2350 1770 FOR I>0 TO HQ 1780 PP«P(I) : GOSUB 2430 1790 FOR J»0 TO 2 1800 IF (1=0) GOTO 1820 1810 C(I,J)=C(I,J)-C(0,J) ; D(I,J)=D(I,J)*D(0,J) 1820 PP=C(I,J) : QQ=SQR(ABS(D(IJ,)>> : GOSUB 2460 1830 NEXT J 1840 LPRINT 1850 NEXT I I860 11=0 : L=l : XX=0t ; YY=0t 1870 LPRINT CHRSI18) : LPRINT SPC(20)CHRS(14)P '; ENETRATION RATIO 1880 LPRINT CHRS(20);' - f s'; : A="-l' 1890 GOSUB 2270 : LPRINT • ]• : GOSUB 2350 1900 FOR I=L TO HQ 1910 PP=P(I) : GOSUB 2430 : BB=(U(I)/S11))'2 : K=0 1920 IF (11=0) GOTO 1940 1930 XX=C(1,X) : YY=D(1X ,) 1940 00=C(I,K)-XX : PP=00/S(I) A 1950 0Q=ABS(PP)"SQR(D<I,KXYY)/002-BB> : GOSUB 2460 1960 IF (X=2) GOTO 1980 1970 LPRINT SPC(28); : K=2 : GOTO 1940 1980 LPRINT 1270  128a 1290 1300  131  1999 HEXT I 2399 11=11-1 : L=2 2313 IF (11=I) GOTO 1373 2029 LPRINT CHR9(1S> : LPRINT S?C(27)CHR9<14);•GRADIENT*; 2333 LPRINT CHRS^);" - t ng.»l*; : A»'-l' : GOSUB 2273 2040 LPRINT •.»••; : GOSUB 2273 : LPRINT • ]• : GOSUB 23S3 2050 FOR 1=3 TO RO 2360 PP=P(I) : GOSUB 2433 2973 FOR J=9 TO 2 2380 PP'G(I.J) : OQ=SOR(ABS(F(I,J)) : GOSUB 2469 2090 NEXT J 2199 LPRINT 2119 NEXT I 2123 LPRINT CHS9UB) : LPRINT S?C( 1S)CHR9( 14) D ;' IFFUSION CONSTANT * ; 2129 LPRINT CHR9I20);* - C cn'; : A="2" : GOSUB 2270 2140 LPRINT "a*; : A»*-l* : GOSU3 2270 : LPRINT • ]• : GOSUB 2280 2150 FOR 1=1 TO SO 2163 PP'P(I) : GGSUB 2430 : BO=G(1,1)-G(I. 3) 2170 LPRINT S?C(23); : PP=C(1,1>/(1000-PP-CC) 2180 00=0(1,i)/C(I, l)"2-.25/P(I)"2-(F(I.l)-F(I,3))/00"2 2190 00= PP-SCRUBS(CO) ) : GOSUB 2463 : LPRINT 2230 NEXT I 2210 ERASE T, W. 3, E. ?, S. U, C. D, 0. N. X, Y, 2, 3. R, G, F, 3, V 2223 QN ERROR GOTO 3 : LPRINT CHRSI12);CNRS<27);CHR9(Sai; "2239 CLGSZ : GOTO 153 2243 LPRINT ChR ' S(12) ;C:-i39(13) ;CHR9(27) ;CHR3(S5) ;CHRS(12); CHRSI27);C4RS<50); 2259 LPRINT S?C(23)*MAnE : •;A£;• C SLOPES - SETina 1' : LPSIHT : LPRINT 2263 RETURN 2273 LPRINT CHR9(27);CHRS(83);CHR9(3); 2283 LPRINT A;CHRS(27);CHR9(84); 2293 RETURN 2330 LPRINT S?C(23)CHRS(27);CKR9(45);CHR9(1);*TIME (am)"; 2213 LPRINT SPCC5P )'OSTERIOR VITREOUS";S?C(9)"SID-VITREOUS;' 2323 LPSIHT 5?C(9)"ANTERIOR VITREOUS';S?C(7)'Points"; 2333 LPRINT CH3S(27);CHRS(45);CHRS(3) 2343 RETURN 2253 LPRINT CHR9I27);CH39(38>;SPC(35)'after •; 2369 IF 11=0 THEN LPRINT "BACKGROUND"; ELSE LPRINT "BOLUS"; 2373 LPRINT • subtraction" 2330 LPRINT CHRS(15) : LPSIHT S?C(201CHR9I27);CHR9(45);CHRs<1); •TISE (nin)"; 2390 LPRINT S?C(5)•POSTERIOR VITREOUS (3 2409 LPRINT S?C(5>S'ID-VITREOUS (9 n/RI'; 2410 LPRINT 3?C(5)A ' NTERIOR VITREOUS (3 an/L)'; CHRS(27);CHRS(45);CHRS(0) 2420 RETURN 2430 LPRINT S?C(22); 2440 LPRINT USING • f».##';PP( s LPRINT SPC(4); 2450 RETURN 2460 LPRINT S?C(S); 2470 LPSIHT USISG •$*.»**"—•;??; 2489 LPRINT • •/-•; 2490 LPSIHT USING •#«.»#»"*«• ;00; 2300 3ETURX 2510 IF HERR>10 THEX 3ESUHE 2213 ELSE NE3R*NE3H»1 2523 LPRINT "SLOPES E3S03 Code «*;£3R;"in Line #';E3L 2530 RESUME NEXT 2540 FOR K=0 TO 4 2550 T(K)=0l : ¥(K)*0t Z"iS0 NEXT X 2570 FOR K=« TO JJ 2 (X)>E(J-1)) CR GOTO 73=00l> GOTO 2729 25 38 90 0 IFIF(X(XIXXE(J) Y(K2X 2639 OX KK GOTO 2620, 2630,2640 2610 BB=X(K) : CC=Y(K) : DD=Z(K)"2 : GO,TO 2559 2623 aa = LCG(X(X)l : CC=Y(K) : DD=ZCX>2 : GOTO 2553 2633 3B»X(X) : CC=LCG(Y(X) : DD-( Z( K)/? (X)) "2' : GOTO 2559 2640 BB-X(!C) : CC=1I/Y(K) : DD=(Z(K)/Y(K)*2)*2 2559 »(4>»lt/DD : ¥(3)"W(4)'CC 2669 FOR L=« TO 4 2679 T(L)=T(L)-»(4> : V(4)*W(4)>BB 2680 NEXT L 2699 FOH L=0 TO 2 2799 »(L)»»(L)-»(3) : *<3)«¥(3>«8B 2710 HEXT L 2729 HEXT I nm/R)';  132  2739 BB=T(1)/TI0> : CC=TC2)/T(1> : DD=TI3)/T(2> : EE=T(4)/T(3> 2749 FF=CC-DD : GG*1!/DD-l1/CC : HH»1I/BB-1!/CC 2750 00=FF-GG-HH"<EE-DD) : PP=W(9)/T(1)-W(1)/T(2) 27S9 Q0=(PP«GG-<W<2)/T(3)-W<1)/T(2) )«HH)/W<3) : V(KK, 41=00, 2779 RR"<PP-0OF .F)/HH : V(KX0 ,)=RR : SS*W(0)/T( 1)-RR/BB-QQ*CC 2789 V(XK.2)=SS : PP=EE-CCA : TT=(EE»<CC-BBA)/CC<-8B-DD-FFABB/DD)A2 2799 UU=EE-DD : QA0=UU2/T (9)-2!*UU»PP-BB2/T(2) 2899 QO=aQA»<ABBP 'AP)2/T(2)-2IABB»FF»UU/(T(9)ADD)*<BBAFF)A2-EE/T(3) 2819 00=<Q02!BB2»PP»FF/T<2))/TT : V(XKI,>"OQ 2829 UU=T(4)/T(A2)-T(2)/T(0>A : FF=DD-3B A 2839 RR»<PP/CC)2/T<0)-UU2/(T(3>«OD)-EE"FF2/T(3> 2849 RR=RR-2!«PAP'UAU"BB/(CC»T<3))«2I•FF»PP«BB/T(3) 28S9 RR»(RR-2!UUFF/T(3))/TT : V(KX3 ,)=RR 2869 PP*DD-CCA : UU=CC-BB : SS= (PP»AT( 1)/T(3) )A2/T(8) 2879 SS=SS*FFA2/(T<3)"DD )->EE('UU/CC) 2/T(3) A 2888 SS=SS-2!(FF-UU) PP"3B«Ttl)/T(3)A2 A A 2899 SS=<SS-2!FFUU/<CC»T!3)))/TT : V<KK5 ,)=SS 2988 IF KK<>2 THEN RETURN 2919 V(XK,3)=EXP(V<KK,a) : V<KK, 1)=V<XX,1)»V<XX,9)A2 2928 RETURN 2939 ON XX GOTO 2958,398A9,2949 2949 CC=PP : EE=1! : FF=2!CC : GOTO 2968 2958 CC=L0G(PP) : EE=A1!/PP : FF=2!»CC-EE 2968 BB=V(XX,9l*V<XX,2)CC-y(XK, 4)»CC*2 2978 IF KX=3 THEN 3B=1!/BB 2988 , IF 11=9 THEN RETURN A 2998 DD=V(KK1,) *Y ( KK, 3)»CC2*V(KK, 5)»CCA4 A 3809 GG=V(KK,2)-EE*V(KK, 4)»FF : HH=V<KK,3)»EE2*V<KK,5>»FFA2 3819 ON KKA GOTO 3030,3070,3040 3020 EE=4IV<KK. 5) : RETURN A 3030 FF=EE»C -GG-FF) : EE = EE2*(HH*41"V(KK, 5)»EEA2) : RETURN 3040 CC=BBA4 : DD=DD«CC : GG=-GG*BBA2 A: HH=HHACC«DD-<2A!-GG/BB)A2 3050 CC=GGB /B : FF=21•(GG»CC-V(KK, 4)«BB2) : EE=16I»HH»CC2 3069 EE = EEA0D<4!»CCA4«-(4t»BBAV<KK, 4))A2>*41»V<KK,5>»BBA4 3070 RETURN 3080 CC=PP : BB=V<KK,0)E ' XP<V(KK,2)-CC»V(KK,4)»CCA2) 3898 IF 11=0 THEN RETURN A 3100 DD = V(KK,1)•(BB/Y(KK, 0))2*V(KK, 3)• ( CC'BB)A2-V(KK,5)•<BB'CCA2)A2 3110 FF=V(KK,2)»2I»V(KK,4)-CC : GG=BB»FF 3120 HH=DD-FFA2*<V(KXA,3)-4I»V<KK,5)ACCA2>»BBA2 :A EE=HH«FFA2 3130 FF=FF«G(WI-V<KK,4)BB : EE=EE*Y(KK,3)"GG2 3140 EE=EE*4I•V(KK,5)A((GG'CC)A2-BBA2>*41"DD"V(KK,4)A2 3150 RETURN  A . 1 3  P L O T . B A S  1800 REN Updated 210186 1018 DIN L(256), .1(256), N<3), SXI15), SYU5), S2!15>, B(3) 1020 A»** — : PRINT : PRINT A : PRINT P ' LOT MENU". : PRINT A 1030 PRINT : PRIHT TAB(191;T ' he files in Drive 3 are:' : FILES'3:*.•' 1040 PRINT : PRINT : PRINT TAB(10):'FILE TYPE" : PRINT TAB(10);A 1059 PRINT "(0) A . VG files -> Averaged rav data" 1060 PRINT "(1) R . ET files -> Retina-aligned, b/g-subtracted" 1070 PRINT "12) C . RP filea -> CRPeak-aligned, b/g-subtracted' 1080 PRINT : PRINT : PRINT W ' hich type do you vish to plot 1090 INPUT I : PRINT : PRINT : PRINT "Choose froii:" 1180 IF (I<3) GOTO 1130 1110 PRINT : PRINT : PRINT "BAD ENTRY! Please try again.' 1120 PRINT : PRINT : PRINT : GOTO 1020 1130 ON I GOTO 1150,1160 1140 AF=".AVG" : K2=0 : FILES 'B:AVG' : GOTO 1170 1150 AF=".RET' : KZ=1 : FILES 'B:RET' : GOTO 1170 1160 AF='.CRP' : KZ'l : FILES 'B:«.CRP' 1170 PRINT : PRINT : N0--1 : PRINT "PRESS RETURN TO EXIT.• : PRINT 1180 LINE INPUT 'Filename -> ";A 1190 IF (A-"> GOTO 1210 1209 HO=MO-l : AN(«0)=A : GOTO 1130 133  1210 OPEN 'I',#l. B ' iSUBJECT. DAT" 1220 INPUT #1,A« : INPUT #1A . : INPUT #1A . : INPUT #1A , 1230 INPUT *1.N<0) : INPUT I I , N i l ) : INPUT #1,N(2> 1240 INPUT #1,3(0) : INPUT #1,3(1) : INPUT #1,3(2) 1250 CLOSE : N(1)=N(1)-N(0> : N(2)»N(2)-N(1) : HV»NO>\2 1260 NA»<HU)*M(2))\2 : PRINT : PRINT : PRINT 1270 FOR 1=0 TO 2 1280 B(I)=B(I)/N(I) 1290 NEXT I 1300 FOR I = KZ TO HQ 1310 SY(2)--13I : SY(3)=>1E*22 : SY(4)=-13I : SY(5)=SY(3) : SY(6)=-101 1320 JJ=-1 : PRINT TAB(10);"Reading ";AN(I> 1330 OPEN "I", #3, "3: "•ANdl'AF : PP=01 : 1)1*6 1340 INPUT #3,P : INPUT #3,0 : INPUT #3,R : K=CINT(P> : JJ=JJ*1 1350 IF (K<>«V) GOTO 1370 1360 SX(0)=P : SY(0)=O : SZ(0)=R 1373 IF (K<>HA) GOTO 1390 13S0 SX(1)=P : SY(l)=a : SZ(1)=R 1390 IF X (>HV OR 0<SY(2)> GOTO 1410 1400 SX(2)=P : SY(2)=0 : SZ(2)=R 1410 IF (K<0 OR X>N<0) OR 0>SY<3)> GOTO 1430 1423 SX(3)=P : SY(3)=0 : SZ(3)=R 1430 IF K (<HV OR X>HA OR Q<SY<4)> GOTO 1453 1443 SX(4)=P : SY(4)=Q : SZ(4)=R 1450 IF (K<N(1) OR K>N(2) OR 0>SY<5)> GOTO 1470 1463 SX(5)=P : SY(5)=0 : SZ(5)=R 1470 IF K (<HA OR 0<SY(6)) GOTO 1493 1483 SX(6)=P : SY(6)=0 : SZ(6)=R 1490 IF X>N(0) THEN GOTO 1530 ELSE SQ=B(0>-P 1533 IF (00<PP) GOTO 1523 1513 IN=I!1«1 : SX(IN)=QO : SY(I!l)=a : SZ(IN)=R : PP=PP-3! 1523 P'3.333-B<0)"P-11! : X=P : GOTO 1583 1533 IF (K>H(D) GOTO 1553 1540 P=3.333»3(1)»(P-N(3))*X : Y=P : GOTO 1583 1553 IF <X>N(2>> GOTO 1573 1563 P=3.333«B(2)-(P-H(11)»Y : Z=P : GOTO 1580 1573 P=.32"(P-N(2))-Z 1580 L(JJ)=CINT(P) 1593 IF Q<=3 .364 THEN S(JJ>=3 ELSE N(JJ)=CINT(. 31831S»(L0G(0)«139.339)"2-212.586) 1630 IF EOF(3)<>0 THEN CLOSE ELSE GOTO 1343 1613 LPRINT CHRS(18);SPC(23)N ' AHE : ";Ai!;" C PLOT ]• : LPRINT 1623 LPRINT : LPRINT : LPRINT SPCI23)CHRS(14);AN(I);CHRSI23); 1633 LPRINT "-minute PROFILE has";JJ-1;"data-seta,• 1643 LPRINT SPC(30)'stored in ";AN(I)»AF;" ." : LPRINT 1650 LPRINT CHRS(151 : LPRINT CHRS(27);CHRS(49) 1660 LPRINT SPC(25)'Vertical (LOG) Scale = X 10 ng/ml per 10 div.'; 1670 LPRINT SPCI7)"Horizontal Scale = 3 mm per 10 div.' 1680 FOR J=0 TO 40 1690 KK=40-J : P«XK/5I : LL=3 : LPRINT CHRSI13);CHRS(13);CHRS(9); 1739 IF ((P-FIX(P))=31) GOTO 1723 1710 LPRINT CHRS19);CHRS1124); : GOTO 1770 1720 IF (KX*5 OR KX=15 OR XX=25 OR KK-35) GOTO 1750 1730 IF J=0 THEN LPRINT 'ng/ml ••; ELSE LPRINT CHRS<9); '*"; 1740 . GOTO 1770 1750 LPRINT" 10" C ; HRS<27) ;CHRS(83) ;CHRS(0) ;XK\10; 1760 LPRINT CHRSI27);CHRS(34);" ••; 1770 FOR X=0 TO JJ 1780 IF (H(XX>KK OR L(X)>110) GOTO 1820 1790 IF L(K)=« THEN Mfl'l ELSE NH=L(K!-LL 1800 IF N«=3 THEN GOTO 1813 ELSE LPRINT SPC(MH-1)CHRS(249); 1813 LL=L(X) 1823 NEXT X 1833 NEXT J 1840 LPRINT : LPRINT CHRSI9);CHRSI9);• •; : LL=0 1850 «M=CINTI.3333»B(0)»N(0))»11 1863 FOR J=0 TO 10 1S70 IF (10»J)=«H THEN GOSUB 2380 ELSE LPRINT 1880 FOR X=0 TO 8 1890 XK=10-J*K-1 1900 IF XX = HI THEN GOSUB 2380 ELSE LPRINT 1910 NEXT K 1920 NEXT J 1933 IF (l««110 THEN LPRINT •'• ELSE LPRINT ••• 1940 LPRINT CHRS(9);CHRS(9);'-3';SPC(9)'R';SPC(19)'6";SPC(18);'12'; 1950 LPRINT SPC(18)"18';SPC(ia)"24';SPC(ia)'mm' : LPRINT : LPRINT 134  1968 1973 1983 1993  LPRINT CHRS(27);CHRS(65);CHRS(12);CHRS(27);CHRS(50> LPRINT SPC(20)CHRS(27>;CHRS(45);CHRS(l);'Landmark'; LPRINT SPC(29)"Concentration -/- S. E. (ng/ml)"; LPRINT SPC(5)"Distance from RETINA (mm)"; CHRS(27>;CHRS<45);CHRS(3) 233a LPRINT SPC(2a>"Choroid-Retinal Peak";SPC(9); : X=B(0)"SX<2) : Y=SY(2) 2013 Z=SZ(2) : G OSUB 2253 : LPRINT "Mid-Vitreous";SPC(17); : X=B(0)'SX(0) 2823 Y»SY(0) : Z=SZ(0> : G OSUB 2258 : X=B(0)»SX(3> : Y=SY<3) : Z-SZI3) 283a LPRINT "Vitreous Lo»";SPC(17); : GOSUB 225a : P»B(0)»N(0> : Y=SY(4) 2a48 LPRINT "Lens Peak";SPC(28); : X=B( 1) MSXM)-N<0) >-P : Z=SZ(4> 2353 GOSUB 2258 : LPRINT "Anterior Lo»";SPC(17); : P'P-B(1)»(N(1)-N(8)) 286a X=B(2)'(SX<5)-K(1))-P :.Y«SY(3) : Z=SZ(5) : G OSUB 225a : Y=SY(1) 2878 LPRINT "Mid-Aqueous";SPC(18); : X=B(2)•<SX(1)-N<1))-P : Z'SZ(l) 2888 GOSUB 2258 : LPRINT "Anterior Peak";SPC(16); : X»B(2)"(SX(6)-N(1))*P 2893 Y=SY(S) : Z=SZ(6) : G OSUB 2253 : LPRINT 213a LPRINT SPC(33)CHRS(27);CHRS(45);CHRS(1); D • ISTANCE from RETINA (mm)"; 2113 LPRINT SPCl18)"READING -7- S. E. (ng/ml)"; CHR9(27);CHRS<45);CHR3(8) 2128 FOR J=7 TO IM 2133 X=SX(J) : Y ' S Y U ) : Z=SZ(J) : G OSUB 2263 2148 NEXT J 2158 LPRINT CHRS(27);CHRS(58);CHRS(12) 2163 NEXT I 2173 PRINT : PRIHT "Enter ESC to exit from plot , • : PRINT 218a PRINT " ANY OTHER key to continue.'; : A=INPUTS(1) : PRINT 2198 IF (A<>CHRS(27)) G OTO 1B28 2238 ERASE L, M, N. SX. SY, SZ, 3 2218 LPRINT CHRS(27);CHRS(65);CHRS(12);CHRS(27);CHRS(53);CHRS(18); 2228 CLOSE : GOTO 158 2233 L PRINT SPC (63) "The data are stored in •; 224a R ETURN 225a KX=0 : G OTO 2333 2268 KX*1 : L PRINT SPC (4a); : GOTO 228a 2278 LPRINT SPCl18); 2288 LPRINT USING •#*.#»*#»•;X; 2293 IF (XK=1) G OTO 2328 2330 LPRINT : LPRINT SPC(28); 231a RETURN 2328 L PRINT SPC(2a); 2338 LPRINT USING •##.#»###«*«*• ; Y ; 234a LPRINT ' •/-•; 2358 LPRINT USING #####****•;Z; 236a IF KK=1 T HEN LPRINT ELSE GOTO 227a 2373 R ETURN 2388 L PRINT : LL=LL-1 2390 MM«CINT(.3333»B(LL)«(H(LL)-N(LL-1) ) 2400 R ETURN *t#.  >»HH  A - 1 4  D R A W . B A S  1000 REM Updated 253186 1313 DM I X(256),Y(256),N(4),Z(4) 1323 A"'-*--' : PRINT A : PRINT D ' RAW : PRINT A : 11=0 1030 ON ERROR GOTO 1970 1040 LPRINT CHR9<15);CHRS(27);CHRS(49);CHRS(27);CHRS(79); : MO*U 1060 FOR J=0 TO HQ 1070 OPEN •I'.fl."BtSUBJECT. DAT' : INP1UT #1A , AA 1080 PRIHT : PRINT TAB(15);N ' ame : A ; AA : PRINT : PRINT 1090 INPUT »1,A : INPUT #1A , : INPUT #1A , llOa INPUT #1K , X : INPUT #1H , S : INPUT fl.LL : INPUT #1,EE 1110 CLOSE : PRINT 'The files in Drive 3 are:' : PRINT 1 3 5  112a  IF u=e> GOTO 115a  zia>»N(a>/KX : Z(i>»<N<i>-N<a)>/HH 1148 Z(2)'(H<2)-NU))/LL : GOTO 1173 1158 NI8XKX : N(1)M ' M-KX : DD=KK/EE : JF=2!«DD-20 11S8 N<2XLLN ' (1> : H<3XKK\2-15 : JG*4!«DD-20 117a LY'-l : MM=HHX *K : LL=LLM *M 118a FILES •Bf.'" : PRINT : PRINT 1198 PRINT TAB(5);"Enter . to change Diskette.' : PRINT 12aa PRINT TABI5);"Enter RETURN to exit." : PRINT : PRINT 1218 LINE INPUT 'Enter the COMPLETE Filename. ';AN(J> 1228 A-RIGHTS(AN(J),4) : PRINT : PRINT 1238 IF (AN(JX""> GOTO 1753 124a IF (AN(J)o'.*) GOTO 12S8 1253 J»J-1 t PRINT : GOTO 1728 126a IF <A="A . VG" OR A*'.RET" OR A»".CRP") GOTO 12B8 1278 PRINT "INCOMPLETE Entry. Please try again.' : GOTO 1188 128a OPEN "I",#2."3:"*AN(J) : NN=-1 1293 NN-NNl' : INPUT #2X . (NN) : INPUT #2, Y(NN) : INPUT #2,ZZ 138a IF XINNX-20! THEN NN=NN-1 1313 IF (NN<0 OR J=0) THEN GOTO 1358 ELSE ZZX ' (NN) 132a IF (ZZ<=LL AND ZZ>MM) THEN X(NN)=(ZZ-MM)Z ' (2!N ' (1) 1338 IF Z (Z<=MM AND ZZ>KK) THEN X(HH)=(ZZ-KX)Z ' (1)N ' <3) 134a IF (ZZ<=KX> THEN X<NNXZZ"Z(3) 1358 IF EOF(2)<>8 THEN CLOSE ELSE GOTO 1298 1363 IF (J<>9) GOTO 1388 1373 LPRINT : GOSUB 186a : LPRINT 1388 MC=8 : L=0 : H=0 : A*RIGHTS<STRS(J),1) 1398 IF N(2)>NH THEN XK*N(2>.10 ELSE KK=NN-18 1438 IF (XX>218) THEN KX=XK«-13 ELSE KK=210 1413 FOR K«0 TO XX 1423 IF (J<>3) GOTO 1538 1433 IF < <N<M)-20X>K> GOTO 1463 1448 IF (M=0 OR H»l) THEN LPRINT "L"; ELSE LPRINT "C; 1453 MH ' -1 : GOTO 1518 1463 IF (XOJF) GOTO 1483 1473 LPRINT "2"; : GOTO 1513 14S3 IF (XoJG) GOTO 1538 1493 LPRINT '4'; : GOTO 1513 1538 IF X<>23 THEN GOTO 1528 ELSE LPRINT "R'; 1513 GOSUB 1863 : GOTO 1533 1523 IF RIGHTS(STRSCK), 1X"0* THEN LPRINT •<••; ELSE LPRINT CHRS1124); 1538 LPRINT CHRSI13);" "; 1548 IF <K<XN(3)-II) OR MC=1) GOTO 1568 1558 GOSUB 1838 : 11=11.2 156a FOR JJ=L TO NN 157a LL=CINT(X<JJ).23I)-X 1588 IF (LL<3) GOTO 1643 1598 IF (LL>8) GOTO 1658 1633 IF Y(JJ)>0! THEN ZZ=L0G( Y( JJ) ) ELSE ZZ=8I 16ia HM=CINTC44.7-25«ZZ». 114»ZZ*2) 1628 IF MM<8 THEN HM=9 1638 IF MM<133 THEN LPRINT SPC<MN)A; 1543 L=L»l : LPRINT CHRSU3);" •; 165a NEXT JJ 1668 L«JJ : LPRINT 1678 NEXT X 1683 IF JM ' O THEN GOTO 1758 ELSE PRINT 169a PRINT TAB(18);"PAUSE";J;"for PAPER Adjustment' 1738 PRINT TAB!12);'(Enter ESC to exit.) •; : A=INPUTS(1) 1718 IF A=CHRS(27) THEN GOTO 1753 ELSE RESET 1723 PRINT : PRINT TAB! 15);•INSERT NEW Diskette."; : A*INPUTS(1) 1738 PRINT : PRINT : PRIHT : PRINT : PRINT : LPRINT : RESET 1 NR EX TT JCHRSI27) C 17 74 58 3 LP IN ; HRS( 65) C ; HRS( 12) C ; HRS< 27); CHRS( 53) ; 1763 LPRINT CHRSI27)C ; HRS(53) : PRINT : PRINT ; A=INK£YS 177a PRINT TAB(13);"Start a NEW plot? (N) •; : A»INPUTS(1) 1788 IF <A=Y ' ' OR A="y"> THEN PRINT ELSE GOTO 1838 1798 PRINT : PRINT : PRINT : J=8 : GOTO 1848 1883 ON ERROR GOTO 8 1813 ERASE X.Y.N.Z 1828 CLOSE : GOTO 158 1838 LPRINT SPC188)"[";RIGHTS(STRS(J),1);•] =•> »;AN(J); 1848 LPRINT • Profile for "A ; AAC ; HRS<131 "j : MC»1 1853 RETURN issa FOR LV»I TO 12a 187a IF RIGHTS(STRS(LV),l)»"a" T HEN LPRINT ELSE LPRINT iaaa NEXT LV 1890 LY-LY'l 113a  136  1909 ON LY GOTO 1920,1930,1940.1950,1950,1960 1910 LPRINT • LOG*; : RETURN 1929 LPRINT • RETINA"; : RETURN 1930 LPRINT • 2 «•'! : RETURN 1940 LPRINT ' 4 mm;' : RETURN 1950 LPRINT • LENS"; : RETURN 1960 LPRINT " CORNEA"; : RETURN 1979 IF (ERL<>1280) GOTO 2019 1980 PRINT : PRINT CHRS(7);TA8(13);"Thfile is does not exist." 1990 PRINT : PRINT TAB!12);"Try again.: "PRINT 01 03 0 PREIN SUTMECH 1R 18S(3 23 7>";ERROR Code #";ERR;"in Line #";ERL 2323 RESUME NEXT  A . 1 5  L U N D . B A S  1339 REM Updated 183236 1910 OM I X(130),YI130), 2(133), 0(100), P(2), 0(2), R(2) 1023 DM I S(2,2>,T(2>,U(2>,V(2,2),'rf(2>,P3(3),PL0(3) 1030 A="--~" : PRINT A : PRINT "LUND" : PRINT A 1340 XP=0 : ASO=CHRS<27>«CHRS(83)a<-CHRS(0)2.'*'CHRS(27)-CHRS(84) 1050 ON ERROR GOTO 3270 : NERR0 1063 OPEN "I",#1."3:SUBJECT.DAT" : INPUT #1A .A 1073 INPUT #1A . : INPUT *1,A ; INPUT #1A , 1383 INPUT #1,X(0) : INPUT #1,Y(0) : INPUT #1,X(2) 1399 INPUT #1,Y(0) : INPUT #1,Y(1) : INPUT #1,Y(2) 1103 CLOSE : RAD=(Y(9)*Y(1)*Y(2))/2! : SC=Y(9)/X(9) 1119 OPEN "I", #2, "9P : LASMA. FIT" : INPUT #2X ,X 1123 FOR 1=0 TO 5 1130 INPUT t2,P0 : J-I\2 1140 IF (1=9 OR 1=2 OR 1=4) THEN PB(J)=P9 1150 NEXT I 1169 CLOSE 1170 PRINT "The files in Drive B are ;• 1130 FILES "B:-.»" : PRINT : PRINT 1190 PRINT TAB(5);"Enter RETURN to exit." : PRINT 1200 PRINT TAB(5);"Enter . to CHANGE diskette." : PRINT : PRINT 1213 LINE INPUT "Enter the COMPLETE ????????.??? : ";AE : PRINT 1220 IF AE<>".• THEN GOTO 1250 ELSE PRINT 1230 RESET : PRINT 'Replace diskette NOW; enter ANY key vhen ready.'; 1240 A-INPUTS(l) : PRINT : RESET : GOTO 1360 1253 IF (AE<>"> GOTO 1290 1260 ERASE 0, P, 0, R, S, T, U, V, W,X, Y, 2, PB, PLO 1270 ON ERROR GOTO 0 : LPRINT CHRSI 12) ;CHRSI27);CHRS(58); 1280 CLOSE : GOTO 150 1290 GOSUB 2340 : PRINT : PRINT 'Enter the LEFT limit •; : INPUT PL 1300 PRINT 'Enter the RIGHT limit •; : INPUT RP : PRINT : PRINT 1310 IF (RP< = PL OR RP>RAD) THEN RP=RAD 1320 LPRINT SPC(19)'Analyzing file : •;CHR9(14);AE;CHRS(20); 1330 LPRINT • vithin t•;PL;',";RP;"1 mm' : LPRINT 1340 A=*INITIAL ESTIMATES of : PRINT 'Enter the ';A 1350 LPRINT SPC(18)A : A"'Permeability Coefficient P (cm/s> =• 1360 PRINT A; : INPUT P(0) : LPRINT SPC(20)A;P(0) 1370 A=I'ncrement of P to be used (cm/a) =' : PRINT A; 1380 INPUT R(0) ; LPRINT SPCI21)A;R(0) : LPRINT SPC(22)"and' 1390 A='Diffusion coefficient D (cm'-ASO-' /s) =• 1403 PRINT 'Diffusion coefiicient D (cmA2/s) ="; 1413 INPUT P(l) : LPRINT SPC(22)A;PU) 1423 A»'Increment of D to be used (cm'-ASO-' /s) =' 1430 PRINT 'Increment of D to be used <cmA2/a> =•; 1440 INPUT R(l) : LPRINT SPCI20)A;R(1) : LPRINT : PRINT 1450 FOR 1=0 TO 1 1460 IF 1=0 THEN P9»600I ELSE P0-6309! 1479 P(I)=P(I)»P9 : R(I)=R(I)-P0 1483 NEXT I 137  1490 PRIST : II = -1 : PP=.001 : AAA=RIGHTS(AE3 , > 15B0 IF LEFTS<AAA2.X>"CV THEN P0=SC ELSE P0=1! lSia O PEN "I",#3,"3:"*AE 1523 IF II>99 THEM GOTO 15S3 ELSE 11=11*1 153a INPUT »3,X(II) : X(II)=P0*X<II) : INPUT *3,Y<II) : INPUT #3, Z(II) 1543 IF X d l X a i THEN 11 = 11-1 1553 IF (E0F(3)=3) GOTO 1523 1563 CLOSE : J=II : II = -1 1573 FOR X=3 TO J 1583 IF (X<X)=>RP) GOTO 1623 1593 IF (Y(K)<=0! OR X(XXPL) GOTO 162a 1633 11=11*1 : X(II)=RAD-X(K) : Y(II)=Y(X) : ZUI)=Z<K)A2 1613 PRINT II*1;"> •;X(II);TAB(13);Y(II);"*/-•;Z(X) 1623 NEXT X 1633 PRINT : PRINT 7AB(10);II*1; 'points vere entered." : PRINT 1643 PRINT : LPRINT SPC(33)11*1;'points vere read. " 1653 FOR J=8 TO 1 1663 T(J)»3I 1673 FOR K=3 TO J 1683 S(J,K)=ai 1698 NEXT X 1730 NEXT J 1718 00=0! : J=LEN(AE)-4 : TT=VAL(LEFTS!AE,J) 1723 FOR 1=0 TO II 1730 GOSUB 2450 : 0I(= )WW : Z1 = Y(I)-WW : 00=QQZ *1A2/Z(I) 1740 FOR 1=0 TO 1 1753 U1 = P!J) : P(J)'U1*R!J) : GOSUB 2458 : U2=WW : P U X U l - R U ) 1763 GOSUB 2453 : U( J) = . 5»(U2-WW)/R( J) : P(J)=U1 1773 NEXT J 1738 FOR J=8 TO 1 1790 TU)=T(J)*Z1"U(J)/ZCI) iaea FOR X=0 TO J 1310 S(J,X)=S(J,X)*U<J)*U(X)/Z(I) 1823 NEXT X 1830 NEXT J 1840 NEXT I 1850 00=00/(11-2) : LPRINT SPC(30)" = > FIRST CHI";ASO.;" =';00 1860 FOR J=0 TO 1 1873 FOR X=3 TO J 1883 S(K,J)"S(J,K) 1893 NEXT X 1900 NEXT J 1918 FOR J=8 TO 1 1923 FOR X=3 TO 1 1938 V(J,X)=S(J,X)/SOR(S(J, J)»S(X,K)) 1943 NEXT X 1958 V(J.J)=11*PP 196a NEXT J 1978 GOSUB 2798 1988 FOR J=a TO 1 1998 W(J)=P(J) 2808 FOR K=a TO 1 2818 W(J)=W(J)*T(K)'V<J,K)/SOR(S(J, J)'S(K.X)) 2B2a NEXT K 2838 HEXT J 2843 GOSUB 2430 : SS=0l : LPRINT 2050 LPRINT SPC(14)CHRS(27);CHRS(45);CHRS(1);"S/No."; 2060 LPRINT SPC<8)"Reading*;SPC(ll)"Fit #1•;SPC(11)"Fit #2'; 2870 LPRINT CHRS(27);CHRS(45);CHRS(0) 2080 FOR 1=0 TO II 2090 LPRINT SPC115); : LPRINT USING "**";I; 2100 LPRINT SPCI10); : LPRINT USING "f#.#•##";Y(I) ; 2110 LPRINT SPC(18); : LPRINT USING •##.####•;0(I); 2123 GOSUB 2450 : LPRINT SPC110); : LPRINT USING "»#.####";W¥ 2130 0(I)=WW : SS=SS*(Y(I)-W*)'2/Z!I) 2140 NEXT I 2150 SS=SS/<II-2) : GOSUB 2400 : LPRINT : PLO(KP)=SS 2160 LPRINT SPC(30)"=> SECOND CHI";ASQ;" =";SS 2170 IF (00=>SS OR KP=2) GOTO 2190 2180 PP=10!*PP : XP=KP*1 : GOTO 1910 2190 FOR J=0 TO 1 2200 P<J)=W<J) :fl(J)»SOR(V(J,J)/S(J,J)) 2210 NEXT J 2220 PP=PP/131 : Pl=P(3)/630! : P2=O(0)/600! : LPRINT 2230 LPRINT SPCl12)'The BEST Fit Values are:' 2240 LPRINT SPC(13) P ' ERHEA3ILITY COEFFICIENT ='; 2250 LPRINT USING •##.##*"AAA"AAA';P1; : LPRINT ' */-'; 2260 LPRINT USING •##.### ';P2; : LPRINT 'cm/a .' 138  2270 Pl=P(l)/6000! : P2=Q(1)/6300! LPRINT SPC(IS)D ' IFFUSION COEFFICIENT = •; LPRINT USING •#•.###""• ;P1; : LPRINT • •/-•; 2300 LPRINT USING •#•.###' P2; : LPRINT 'cm';ASQ; Vs .' 2310 LPRINT SPCC20)"with Reduced CHI';ASQ;' =•;SS;CHRS<12) 2320 IF (SS>2!) GOTO 1910 ELSE GOTO 1170 2330 REN Subroutine to print page title. 2340 LPRINT CHRSU8) ;CHR9(27) ;CHRS(65) C ; HR9<12) ;CHRS(27) C ; HR9<53>; 23S0 LPRINT SPC(25)N ' AHE : •; AA; • C LUND ]• 2360 LPRINT : LPRINT SPCI15)'Radius of retinal curvature CHR9I247);' 2370 LPRINT USING •»#.###«• ;RAD; : LPRINT • m m .• : LPRINT : LPRINT 2380 RETURN 2390 REM Subroutine far exchanging curve-fitting parameters, P(I). 2400 FOR J = 8 TO 1 2410 SWAP P(J),W(J> 2420 NEXT J 2430 RETURN 2440 REM Subroutine for calculating function, 0(1). 2450 P0=SQR(P(1>) : P1=P(0)/P0-P0/RAD : P6=P(0)"RAD/(X(I) *P0) 2460 P2=.56419»PS : P3=P6-P1 : P5 = . 5/P3 : P4=P5-(RAD-X(I) ) 2470 P5=P5-(RAD-X(I)) : P6=P3«EXP(2!•P1'P4) : P7=P3-£XP(2! »P1«P5) 2480 WW=8! : T2 = 3! : N=INT(TT) 2490 FOR JI*1 TO N 2500 TU=JI : GOSUB 2A590 : Tl = l!/SCR < TU) 2510 0F=P2»T1-<EXP(-P42/TU)-EXP<-P52 ' /TU)) 2520 0D=EXP(TU«P1*2) : P3=P1/T1 : 01=P3-T1»P4 2530 GOSUB 2710 : 0F=aF-?6'QE»aD : 01=P3-T1»P5 2540 GOSUB 2713 : 0F=0F*?7»QE»0D : QF=TPQ 'F 2550 WW=WW(T . 2*0F)/2! : T2-QF 2563 NEXT II 2570 RETURN 2583 REM Subroutine for calculating plasma value, TP. 2593 T1=TT-TU 2638 IF Tl=3! THEN TP=3! ELSE GOTO 2623 2513 RETURN 2623 ON XX GOTO 2643,2638,2650 2533 GOTO 2653 2643 T1=L0G(T1> : GOTO 2653 2653 T1=1!/T1 2663 TP=PB(3)'P3(1)»T1*PB(2)*T1*2 2673 RETURN A 2683 TP=PB(8)EXP<P9(1)AT1-PB(2)AT1A2) 2698 RETURN 2738 REM Subroutine for calculating error function, OE. 2713 01=01/58! : QE=31 : 02=1! : 04=01/2! 2720 FOR K=l TO 58 A 2733 SK=X : P0=<SK»Q1)2 : P0=EXP(-P8) 2740 QE=QE»04M02*P3) : 02=P3 2758 NEXT X 2753 0E=l!-1.1283a»0E 2778 RETURN 2733 REM Subroutine to inverting matrix, V(I,J), and find det., 00 2798 DM I IA(2),JA(2) 2338 30=1! 2813 FOR X=8 TO 1 2820 01=0! 2830 FOR I=X TO 1 2840 FOR J=K TO 1 2850 IF (A3S(01)>ABS(V(I, J)>) GOTO 2870 2360 01=V(I,J) : IA(X)=I : JA(K)=J 2873 NEXT J 2880 NEXT I 2893 IF (Ol<>0!) GOTO 2910 2900 00=01 : GOTO 3250 2910 IF (lA(XXK) GOTO 2830 2923 IF (IA(K)=K) GOTO 2960 2930 FOR J=0 TO 1 2943 S1=V(K,J> : V(K,J)=V(IA(K),J) : V(IA(K),1)=-Sl 2950 NEXT J 2963 IF (JA(KXK) GOTO 2330 2970 IF (JA(X)=K) GOTO 3010 2980 FOR 1=0 TO 1 2990 S1 = V(I,K) : V(I,X)=V(I, JA(K)) : V( I, JA(X))«-Sl 3003 NEXT I 3018 FOR 1=0 TO 1 223a 2290  139  /  3020 IF I<>K THEN V(I,X)=-V(I, K)/Q1 3030 HEXT I 3040 FOR 1=0 TO 1 3050 FOR J=0 TO 1 IF (K>K AND J<>8) THEN V(I, J)=V(I, J)*V(I, K)"V<K,J) 3060 NEXT J 3070 3080 HEXT I 3090 FOR J=0 TO 1 3100 IF K>K THEN V(K. J)=V(K, Jl/01 3110 NEXT J '1 3120 Y(K,X)=1!/01 : 00=Q0Q 3130 NEXT X 3140 FOR J=0 TO 1 3150 K=l-J 3160 IF (IA(KX = K) GOTO 3200 3170 FOR 1=0 TO 1 3180 S1=V(I,X) : V(I,X)=-V(I, IA(K)) : V(I, IA(X))»S1 3190 NEXT I 3200 IF (JA(KX = K) GOTO 3240 3210 1=0 TO 1 3220 FOR S 1 = V (K,I) : V(K,I)=-V(JA(K),I) : V(JA(X),I)=S1 3220 TI 3240 NEXTNEX J 3250 ERASE IA.JA 3260 RETURN 3270 IF NERR>10 THEN RESUME 1260 ELSE NERR=NERR1* 3230 PRINT CHRS(71";ERROR Code #";EHR;"in Line #";ERL 3290 RESUME NEXT  A .  I S F  i  l  -  F o r m a t s  a) Filenames ??.DAT Created by VITSCAN. BAS Format : (1) Pod "zero" position at beep (in DAS units). (2) Pod position (in DAS units). (3) R/H-Log Amp reading (in DAS units). (4) Repeat (2), (3) ??.AVG ; ??.RET ; ??.CRP b) Filenames Created by REDUCE. BAS Format : (1) Retina-zeroed Pod position (in DAS units). (2) Averaged Log Amp output (in*'). (3) Standard deviation ot (2) (in"). (4) Repeat (1),(2),(3) Filename Created by(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13)  SUBJECTD . AT SCANMsENn Ua.m BA Subject' e.S Subject's age. Eye scanned. Date oi scan. Vitreal length (in DAS units). Lens thickness (in DAS units). Aqueous depth (in DAS units). Ultra-sound vitreal length (in mm). Ultra-sound lens thickness (in mm). Ultra-sound aqueous depth (in ma). Fluorescein injected (in ml). Coaaents and observations. Repeat ii necessary, (12). 140  d) Filename Created by Format : (1) (2) (3) (4) (5) (6) (7)  PLASMA. DAT : PLASCAN. 3AS Background sanple time (aet » 0). Background average (in DAS units). Background standard deviation (in DAS unita). Blood sampling time (in minutes p.i.). Average fluorescence reading (in DAS units). Standard deviation (in DAS units). Repeat (4),(5),(6)  e) Filename Created by Format : (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11)  PLASMA. FIT BLOOD3.AS Best-fit function's code number. A coefficient. Error of A. 3 coefficient. Error of 3. C coefficient. Error of C. Measurement p.i. time (in m inutes Area up to (8) (in"1.min).p.i.). Error of (9) (in'. min). Repeat (8),(9),(10)  f) Filenames Created by Format (1) (2) (3)  : ??.CV? : LUNDB .AS Position from Retina (in mm). 1 Average concentration (in"). Standard deviation of (2) (in Repeat (1),(2),(3)  (4)  141  •7  S a m p l e  C O N S E N T  THE  F O R M  UNIVERSITY OF BRITISH COLUMBIA VANCOUVER. B.C.. CANADA V5Z 3N9  FACULTY OF MEDICINE DEPARTMENT OF O P H T H A L M O L O C Y I1S0 WILLOW STREET TELEPHONE >7S-M*l L O C A L 1431  CGNSEMT_FgRM lb? D?^ei2Bn!?Di of a Vitreous EIy9r9phgtgmeter Studying Alterations in the 31ggd Retinal Barrier Dr. I.S. Begg ; Dr. T. 'Cox; Dr. D.A. Balzarini; Mr Pang K. T. A study is being carried out to measure the abnormal leakage from retinal blood vessels which indicate cell damage in the early stages of disease prior to visible retinal changes. In most patients and normal sub jects, the measurements will be carried out following fluorescein angiography (photography) which is a customary diagnostic . procedure frequently used in clinical practice to visualize damage to the retinal structure, and to indicate disease severity which is useful in prognosis and in planning treatment. The measurement of fluorescein leakage is made in a follow-up procedure called VITREOUS FLUOROPHOTOMETRY at several time intervals fallowing the angiography. The vitreous cavity of the eye is scanned by the light beam <of an adapted clinical microscope,) which is directed into the eye through a contact lens. The scan of an eye takes about 30 seconds with minimal discomfort. After each scan, a finger-prick blood sample is taken to measure plasma fluorescein. There are no side effects related to these measurements. The injected fluorescein dye colours the skin slightly yellow for about 4 hours, and urine for about 24 hours. About 47. of patients experience brief spells of nausea. Serious allergic reactions are rare and have not been encountered in over some 20000 fluorescein angiograms done in the Department of Ophthalmology, UBC. The results of the tests and the personal medical records will be kept confidential by using a code number for each patient. The entire procedure, with repeated scanning, takes about 90 ainutes. The test may only be carried out with your signed consent and the understanding that you may decline to participate, or withdraw at any time during measurement without jeopardizing any routine medical treatment. CONSENT for the procedure and acknowledgement of receigt of § 992Y. of the consent fgrm^ Signature of. Patient  Date  142  Signature of Witness  A P P E N D I X  J3  M A T E R I A L B.  1  U S E D  E l e c t r o n i c s  TYPE  QTY  Capacitors (in nF) 1000 30 10 0.15 10000 100  REMARKS (Code CI C2 C3 C4 C5 C6  -  Colour & Shape) red box yellow cylinder grey/yellow box violet-white blue/yellow bulb orange  Analogue-to-Digital Converter ADC0804LCN 1  ADC  -  black 20-pin  Analogue Multiplexer IH6108  1  MUX  -  black  Operational Amplifiers TLD81ACP 2  OP  -  black 8-pin  2  S/H  -  violet-gold  1 1  PI P2  -  black h a l f - c y l i n d e r black w/a heat sink  2 3 1 1 3 1 1 1 2  Rl R2 R3 R4 R5 R6 R7 R8 R9  -  grey trim-pot grey trim-pot grey trim-pot grey trim-pot cylinder cylinder cylinder cylinder cylinder  Sample-and-Hold IH5111IDE Voltage  Regulators LM336 2. 5 V LM340T5 5.0 V  Resistors (in kOhms) 10 20 50 100 10 1.8 1 0.18 0.1  Table 23.  2 3 2 1 3 1  16-pin  Electronic components of blue c i r c u i t of the DAS. (See Figure 12.) 143  16-pin  board  B. 2  Equipment  1. LOGARITHMIC AMPLIFIER Model E97 (built by the Electronic Shop, Department of Physics). 2. 15-V 100-mA POWER SUPPLY (designed by the Electronic Shop, Department of PhyBics). 3. GAMMA SCIENTIFIC DIGITAL RADO IMETER Model DR-2. 4. GAMMA SCIENTIFIC PHOTOHULTIPLIER DETECTOR Model D-47A. 5. GAMMA SCIENTIFIC SCANNING PHOTOMETRICM . ICROSCOPE EYEPIECE Model 700-10-30X (Left ocular). 6. GAMMA SCIENTIFIC SCANNING PHOTOMETRC I MICROSCOPE EYEPIECE Model 700-10-34A (Right ocular vith fibre optic). 7. Modified NIKON ZOOMP -HOTO SLIT LAMP MICROSCOPE. 8. 2-V VOLTMETER (built by Stephen CLARK to monitor the lamp intensity). 9. A MODEL, EYE (built to required specifications by the Machine Shop, Department of Physics). 19. SPECTROTECH FILTERS: a. SE4 - excitor filter. b. SBS - barrier filter. 11. KEPCO POWER SUPPLY Model RMK 09-S (for slit lamp). 12. OSBORNE 1 64K Microcomputer. 13. OKIDATA MICROLINE 192 Dot Matrix Printer. 14. HAYES SHARTMODElf 300 (for communicating with UBCHet). 15. FISHER ACCUMET Expanded Scale Research pH METER Model 320 (for preparing and measuring buffer pH). IS. IEC CENTRIFUGE Model CENTRA-4 (for spinning blood samples). 17. HAMLITON MICROLITER #702 Hicropipette (for measuring out plasma samples). 18. SONOMETRC I Ultrasonic Digital Biometric Ruler Model DBR 400 (for measuring intra-ocular lengths). 19. CONTACT LENSES: a. Piano PERHALENSR Hydrophilic Contact Lens (soft lens). b. COOPER VISION Plano-concave Hard Plastic Lena. 20. RED TIP HEPARINIZED MICRO-HEMATOCRIT CAPILLARY TUBES (for collecting blood saaples). 21. FUNDUSCEIN Fluorescein Sodiua 25X Aapoules (for intravenous injections). 22. BETHOCEL 2X Sterile (vater-based, highly viscous methylcellulose for holding the hard lens in place, and, to provide the optical continuity betveen interfaces). 23. CYCOLGYL' IX (Cyclopentolate Hydrochloride for dilating the pupil, and anaesthetizing 144 the cornea). 24. M 4 3 Fluorescein Sodiua povder (for aaking calibration solutions). 25. MONOPAN 200 g Balance (for preparing calibration solutions).  B . 3  M o d e l  E y e  The cross-section of the model eye i s shovn below.  The  middle  component i s made of l u c i t e (or p l e x i g l a s s ) . A l l other components are of aluminium. A l l dimensions are i n mm. by four bolts <horizontal dash l i n e s ) . 145  The assembly i s held  together  A P P E N D I X C A L I C . 1  CZ  B R A T I O M  R E S U L - T S  P o d - D A S The least-squares f i t was to the straight l i n e , Y  by setting  =  A  +  B *X  X  =  Osborne/DAS units, and  Y  -  translation i n mm .  The gradient of the f i t was found to be 0.095919  0.000070  mm per DAS unit.  The correlation c o e f f i c i e n t was 0.999.  C A L I B R A T I O N S  O F  S L I T  L A M P  T R A N S L A T I O N  40-  0SB0RNE/D.A.S. OUTPUT UNITS Figure 31.  Calibration curve of the Pod-DAS. 146  DISPLACEMENT  OUTPUT/DIFFERENCE  0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 29 30 31 32 33 34  Table 24.  The mm,  8 16 23 31 38 45 52 59 66 73 80 86 93 100 107 113 120 127 133 139 146 152 159 165 172 178 185 191 197 203 209 215 221 228  8 7 8 7 7 7 7 7 7 7 6 7 7 7 6 7 7 6 6 7 6 7 6 7 6 7 6 6 6 6 6 6 7  Results of Pod c a l i b r a t i o n s .  data under the DISPLACEMENT column are multiples of 0.635  i . e . the pod vas advanced 1/40'  h  147  of an inch at a time.  C . 2  L o g a r i t h m i c  A m p l i f e r  The least-squares f i t was to a logarithmic function, Y  =  A  +  B • log X  where input voltages, X were i n mV and output voltages, Y were i n V. The  c o r r e l a t i o n c o e f f i c i e n t of t h i s f i t was 0.997. INPUT VOLTAGE <mV)  OUTPUT VOLTAGE (V)  1.2 1.5 2.1 2.4 2.5 4.5 5.7 7.9 11.2 17.5 23.4 32.2 46.8 57.5 70.0 99.8 109.0 180.0 224.0 343.0 400. 0 489.0 590.0 778.0 878.0 1000.0 1245.0 1574.0 1923.0 2230.0 3142.0 4085.0 5091.0 6000.0 7000.0 9000.0 Table 25.  0.40 0.50 0.65 0.70 0.73 1.00 1.11 1.25 1.40 1.61 1.73 1.87 2.04 2.12 2.22 2.37 2.42 2.63 2.73 2.79 3.99 3.08 3.16 3.28 3.33 3.38 3.50 3.60 3.68 3.75 3.91 4.02 4.12 4.20 4.27 4.38 Results of Log Amp test. 148  0-j  :  :  : i • , ; l..  1  10  • . . i l ...  :  :  :  100  :  r  i. : ;  1000  10000  INPUT VOLTAGE in mV Figure 32.  C.3  p H  Performance of the Log Amp.  D e p e n d e n c e  Water X  Buffer Y  6. 7 47 67 78 105 115 195 630 1830 3250 3400 8900  Table 26.  X  35.5 82 97 112 105 105 120 152 172 202 193 205  12.4 39 47.5 - 69.5 84 103 142 410 870 1220 4850 9258  Y 70 100 103 115 118 126 143 154 179 178 219 228  Water and buffer sample differences. 149  To test the pH dependence mentioned  i n Section 3.1, tvo sets of  sample solutions vere prepared. The f i r s t set vas made v i t h  deminera-  l i z e d d i s t i l l e d vater; the second set vas made v i t h the pH 7.4 Sorensen's  buffer solution.  (See Section 4.1.) Note that the buffer set  vas not the f i n a l calibration set given belov i n Appendix C.4. The results of linear least-squares f i t ,  of concentrations, X  ( ) to DAS outputs, Y (Osborne/DAS units), vere: 1  Y  = -7.354  • 24.449 • In X  f o r vater samples,  Y  = 13.347  +  f o r buffer sample.  23.897 * In X  The  correlation coefficients vere both 0.99. The slopes of the tvo  fits  vere not s i g n i f i c a n t l y different f o r a t(12-2) test (P = 5%).  The  "intercepts"  vere  s i g n i f i c a n t l y different  (P =  0.5X) Hence,  buffer samples produced higher outputs than vater.  300-i  00  250  3  200  X  Q_  AA  — I O CO  AX  150  < S  cc  1 0 0  X  ^  Legend  O CD  in O  A BUFFER X WATER  50 0H .  1—  i  1—i  i 111HI  io  1—i  100  i 111MI  1  i  1000  i 1111M  10000  SAMPLE CONCENTRATIONS in ng/ml  Figure 33.  pH dependence. 150  i  '  100000  C . 4  R / M - L o g  A m p - D A S  F i t s t o two f u n c t i o n s were done. They were: a)  X = A + B » l o g Y + C »  b)  Y  =  A » exp< B • X  X  =  Osborne/DAS averaged r e s u l t , and,  Y  =  sample c o n c e n t r a t i o n i n"  where  Fit  <a)  was i n v e r t e d ,  Qsborne/DAS  output.  deviations  (especially  chi-square  •  <log Y)*  C • X" )  , and,  ,  1  .  and then compared t o f i t (b) a t  F i t (a) was chosen because i t produced a t the lower  concentrations).  The  smaller reduced  was 2.765. The c o e f f i c i e n t s were A = 4.47039 +/- 1.65113 ,  B = 25.0257 +/- 0.73117 , and, C = 0.11444 +/- 0.06722 . The  every  r e s u l t i n g equation on i n v e r t i n g Y  =  exp( AA  +  J  X/CC  *  IF  f i t ( a ) , was  ) ,  where AA = -109.339 , BB = 11915.9 , and, CC = 0.11444 .  RADIOMETER/LOG AMP CALIBRATIONS  30  100  130  200  230  0SH0RNE/0.A.S. OUTPUT UNITS  F i g u r e 34.  Calibration  c u r v e f o r R/M-Log Amp-DAS.  151  Table  Note This  X  Y  6357.47  231.34  1.07  0. 46  5487.33  227.51  0.92  0. 40  2143.54  204.83  -1.67  -0.82  719. 50  174.66  -0.61  -0. 35  313.79  152.80  -0.68  -0.45  143.01  134.50  -3.01  -2. 24  30.41 30.41  90. 26 90. 58  1.00 0.68  1.11 0.75  29.87 29.87  87.95 90.09  2.85 0. 71  3.24 0. 79  20.42 20. 42  80.17 80.25  0.84 0.76  1.05 0.94  13.78 13.78  70.08 70.37  0.81 0. 53  1.16 0.75  7.52 7. 52  58.24 57. 32  -2.81 -1.89  -4.82 -3. 30  27.  Fit-Y  X  Calibration r e s u l t s of the R/M-Log Amp-DAS.  that several scans were made at the lower concentrations.  was to "weight" the lower part of the c a l i b r a t i o n curve  during  curve-fitting.  The reason f o r doing t h i s was to "improve" the c a l i -  bration  i n t h i s region so that PMT and R/M noise  curve  calibration  procedure  during the  would not adversely affect the c u r v e - f i t t i n g  process. 152  C . 5  A -fc. -h e n u a -fc. dL o n  The e f f e c t s of the attenuation was studied during the t i o n s of the R/M-Log Amp-DAS.  calibra-  Scans through the sample c e l l s holding  various sample concentrations were made. Figure seen.  The  solution.  36  shows two samples where  attenuation  effects  were  peak was the p o s i t i o n when the e n t i r e diamond was i n  the  Attenuation  translation  axis  of  effects  were  each scan at the  studied peak;  by  re-zeroing  the  then  estimating  the  (negative) slope ( i f i t e x i s t e d ) . Refraction different  caused  the  from the diamond,  displacement of the s l i t lamp l i k e the F-numbers ( i n Figure  to 3).  be The  conversion i s given below.  Consider a p r o f i l e where s i g n i f i c a n t attenuation was The negative,  concentration-dependent  slope,  p l o t i s found (from Eq. 5, Section 2.2) to be 153  observed.  B(c), i n the semi-log  Figure 36.  Attenuation i n sample solutions. 154 •  i  c where  c  =  A(c)  -  B(c) * X ,  and X are the concentration and  A(c) i s a "constant".  translation  B(c) represents the decrease i n log-concentra-  tion per unit increase i n the distance the probe focus When B(c) i s small, of  the  respectively.  there i s l i t t l e attenuation.  "penetrates".  Hence, an estimate  concentration at which B(c) i s small approximates the  lower  l i m i t above which attenuation should be taken into consideration. The table  below  gives  the r e s u l t s of the  straight-line  fits  to the  p r o f i l e s and t h e i r respective c o r r e l a t i o n c o e f f i c i e n t s .  c"  B(c)  1  9493 6357 5487 2143  Table 28.  mm"  r  1  0.994 0. 999 0.999 0.982  0.289 0.240 0.182 0. 055  Concentration and gradients of attenuated samples.  A least-squares f i t to B<c)  =  AA  +  BB * c  was made. The results were AA = 1.08*10-  and BB = 3.25*10"  3  5  ; r = 0.97 .  Hence, f o r B(c) = 0, c = -33* ! From Eq. 5, f o r a mean aqueous chamber depth of d(=X) = 3.5mm, at c  ft  = 1000" , 1  B(c") = 0.034,  attenuation < 12% ; and,  at c" = 100" , 1  B(c*) = 0.0043,  attenuation < IV. . 155  C.£5  P e r f o r m a n c e  D a t a  UNITS  CHARACTERISTIC/PARAMETER  S l i t width  0. 1  mm  S l i t height  2  mm  Probe diameter  0. 45  mm  16  Beam-Probe angle Lamp intensity monitor F i l t e r overlap at 502.9nm Max.  s l i t lamp displacement  Max.  i n v i t r o concentration  o  141(1)  mV  0.5  7.  24. 5  mm  6000"  1  AR i n v i t r o  3  AR r a t i o i n vivo on Normals  0.032  LLoD i n vivo  4. 4"  0. 77  ng. ml"* Osb"  S e n s i t i v i t y at 2000"  85. 16  ng. ml" Osb"  EoM at 20"  -0.03  S e n s i t i v i t y at 20"  1  1  1  EoM  at 2000"  1  1  1  1  0. 06  1  R i n vivo  19  No. of data-points i n 10 s  Table 29.  mm  800  % pairs  Performance c h a r a c t e r i s t i c s of the VF system. S e n s i t i v i t y and EoM are i n v i t r o estimates using the c a l i b r a t i o n s in Appendix C.4. Note: S e n s i t i v i t y i s defined as gradient of c a l i b r a t i o n curve at given concentration. 156  . "7  M o d e l  Eye?  CONCENFRfirION  o  S«=san PROFILE  P r o f i l e I NT H E MODEL  ETE  <>•>  C/R  AQUEOUS  VITREOUS 9.03E-6 0.0  I 4.0  4=. 0 9 7 E - 7 , 8.0  1 . 13E-6 , 12.0  , 16.0  1.01E-7 , 20.0  j _ 24.0  TRANSLATION IN MM Figure 37.  Model eye sample profile.  157  A P P E N D I X  O  G L O S S A R Y  BLOOD-RETINAL BARRIER  BOLUS  -  -  The b a r r i e r that separates the r e t i n a l neural tissues from the blood.  A concentrated mass of pharmaceutical preparartion given intravenously f o r diagnostic purposes; a mass of scattering material.  CHOROID  -  The network of small blood vessels immediately behind the retina.  DIABETIC RETINOPATHY  ENDOTHELIUM  -  -  A non-inflammatory disease of the r e t i n a due to diabetes vhich can lead to blindness.  The layer of e p i t h e l i a l c e l l s that l i n e s the c a v i t i e s of the heart and of the blood and lymph vessels, and the serous c a v i t i e s of the body.  EPITHELIUM  -  The covering of the i n t e r n a l and external surfaces of the body, including the l i n i n g of vessels and other small c a v i t i e s . It consists of c e l l s joined by small amounts of cementing substances.  EMMETROPIC  -  When rays entering the eye p a r a l l e l to the optic axis are brought to focus exactly at the retina.  FENESTRATED FUNDUS  -  -  That part of the back of the eye furthest from the pupil.  HAEMOGLOBIN  -  HAEMACYTOMETER HEMATOCRIT  Pierced with one or more openings.  -  The red pigment of the blood carried by the red blood c e l l s ; i s composed of globin (a protein) and haem <an iron compound); i s the means of oxygen transport. -  Instrument f o r counting blood corpuscles.  Volume X of erythrocytes i n whole blood. O r i g i n a l l y applied to the apparatus of measurement.  Many of the above descriptions were summarized from "The Penguin Medical Encyclopedia" by Peter Wingate (U.K. 1976), or, from the "Dorland's I l l u s t r a t e d Medical Dictionary", 26th ed., published by W.B. Saunders Co., (Toronto, 1981). 158  HOMEOSTASIS  -  The primary function of most organs; the processes of maintaining constant physical and chemical conditions v i t h i n the body despite external changes.  HYPERTENSION  -  High blood pressure secondary to s p e c i f i c disease, or, e s s e n t i a l .  INFILTRATE  -  To penetrate the i n t e r s t i c e s of a tissue. Material deposited by i n f i l t r a t i o n .  JUNCTIONAL COMPLEX  LESION  -  -  -  -  Bulge at the weak point i n the wall of an artery.  MULTIPLE SCLEROSIS  OPTIC DISC  -  PERIPHLEBITIS  SEQUELA  Surfaces of the cavity or channel within a tube or tubular organ.  Usually the r e t i n a l macula. In general, any area that i s distinguishable from i t s surrounding by colour, etc.  MICROANEURYSM  PLASMA  The i n t e r c e l l u l a r arrangement between the adjacent columnar e p i t h e l i a l c e l l s .  Any pathological or traumatic discontinuity of tissue or loss of function of a part.  LUMINAL SURFACES MACULA  -  -  -  A chronic central nervous system disease in which the nerve f i b r e s lose t h e i r protective myelin sheaths and t h e i r a b i l i t y to conduct impulses.  The intraocular portion of the optic nerve formed by f i b r e s converging from the r e t i n a and appearing as a pink to white disc. -  Inflammation of the tissues around a vein, or of the external coat of a vein.  Sticky, pale amber l i q u i d with f a i n t , s i c k l y smell; the medium i n which v i t a l substances are transported to a l l body tissues; a solution i n water of salts, proteins, glucose, etc. -  Any lesion or a f f e c t i o n following or caused by an attack of disease.  ZONULAE OCCLUDENS  -  That portion of the junctional complex of columnar e p i t h e l i a l c e l l s , just beneath the free surface, where the i n t e r c e l l u l a r space i s obliterated. It extends completely around the c e l l perimeter. Also c a l l e d •tight junctions".  159  A P P E N D I X  E  A B B R E V I A T I O N S  U S E D  A ADC AR AVG  Operational Amplifier; Figure 12. Analogue-to-Digital Converter; Figure 12. Axial Resolution. Averaged data (by RET); f i l e t y p e .  BAB BRB b  Blood-Aqueous Barrier. Blood-Retinal Barrier. attenuation or extinction c o e f f i c i e n t ; Eq. 5.  CR CRP c  Choroid-Retina(l). Alignment by CR peaks; f i l e t y p e . concentration ( ); superscripts f o r various uses.  D DAS DAT DRP DVM d  Diffusion constant; superscript. W means i n water. Data Acquisition System. (Raw) Data ( f i l e t y p e ) . Diabetic Retinopathy. D i g i t a l VoltMeter; Figure 8. Usually depth, or distance, or displacement.  EoM  Error of Measurement.  H« H" hex  Null Hypothesis. Alternative Hypothesis. hexadecimal (base-16) number(ing).  LLoD Log Amp  Lower Limit of Detection. Logarithmic Amplifier.  MS MUX  Multiple Sclerosis. MUltipleXer (analogue e l e c t r o n i c switch); Figure 12.  P PMT PR3 PR3» p. i .  1  Probability of Type I error - rejecting a true H'; Permeability constant; permeability index i f with superscript I. PhotoMultiplier Tube. Penetration Ratio averaged about 3mm from retina. Penetration Ratio at 3mm from retina, post-injection. 160  R RET RPE R/M  Reproducibility; Eq. 27. Alignment by v i s u a l l y pin-pointed RETinae; f i l e t y p e . Retinal Pigment Epithelium. RadioMeter.  S.D.  Standard Deviation.  S/H  Sample-and-Hold electronic chip; Figure 12.  t >  F i r s t blood sampling time.  VF  Vitreous Fluorophotometry  (1  161  (Fluorophotometric).  R E F E R E N C E S  1.  D.M.  Maurice: "A New Objective Fluorophotometer", Exp. Eye Res., 2, 33 (1963).  2.  S.R. Waltman, H.E. Kaufman: "A New Objective S l i t Lamp Fluorophotometer", Inves. Ophth., 9, 247 (1970).  3.  J.G. Cunha-Vaz: "Sites and Functions of the Blood-Retinal Barriers", i n The_Blggd2Retinal_Barriers, J.G. Cunha-Vaz, ed., Plenum Press, N.Y. (1980).  4.  J.G. Cunha-Vaz, J.R. Faria de Abreu, A.J. Campos, G. M. Figo: "Early Breakdown of the Blood-Retinal Barrier i n Diabetes", B r i t . J. Ophth., 59, 649 (1975).  5.  T.C. Prager, H.H. Chu, C. A. Garcia, R.E. Anderson, J.B. Field, E.A. Orzeck, J.P. Comstock: "The Use of Vitreous Fluorophotometry to Distinguish between Diabetics with and without Observable Retinopathy: Effect of Vitreous Abnormalities on the Measurement", Inves. Ophth. Vis. S c i . , 24, 57 (1983).  6.  J.G. Cunha-Vaz, R. Zeimer, P. Mahlberg, H. Tessler: "Vitreous Fluorophotometry Studies i n Pars P l a n i t i s " , in Retinal_Diseases, Grune & Stratton, Inc., N.Y. (1985), pp. 53-7.  7.  J.G. Cunha-Vaz, C C . Mota, E.C. Leite, J.R. Abreu, M. A. Ruas: "Effect of Sulindac on the Permeability of the BloodRetinal Barrier i n Early Diabetic Retinopathy", Arch. Ophth., 103, 1307 (1985).  8.  T. Engell, 0.A. Jensen, L. Klinken: " P e r i p h l e b i t i s retinae in multiple s c l e r o s i s . A histopathological study of two cases", Acta Ophthal., 63, 83 (1985).  9.  A.C. Arnold, J.S. Pepose, R.S. Hepler, R. Y. Foos: "Retinal P e r i p h l e b i t i s and R e t i n i t i s i n Multiple S c l e r o s i s : 1. Pathologic Characteristics", Ophthal., 91, 255 (1984).  10.  T. Engell, P.K. Andersen: "The Frequency of P e r i p h l e b i t i s Retinae i n Multiple Sclerosis", Acta Neurol. Scand., 65, 601 (1982).  11.  B.R. Younge: "Fluorescein Angiograpghy and Retinal Venous Sheathing i n Multiple Sclerosis", Can. J. Ophth., 11, 31 (1976).  162  12.  C.R.  Bamford, J.P. Ganley, W.A. Sibley, J.F. Laguna: "Uveitis, Perivenous Sheathing and Multiple Sclerosis", Neurology, 28, 119 (1978).  13.  T. Engell, A. Hudberg, A. Uhrenholdt: "Multiple S c l e r o s i s : P e r i p h l e b i t i s R e t i n a l i s et Cerebro-Spinalis - A Correlation between P e r i p h l e b i t i s R e t i n a l i s and Abnormal Technetium Brain Scintigraphy", Acta Neurol. Scand., 69, 293 (1984).  14.  R.C.  15.  J.R. Gray, M.A. Hosier, B.M. Ishimoto: "Optimized Protocol for Fluorotron" Master", presented at the International Meeting on Ocular Fluorophotometry, Paris, France (1982).  16.  T.C. Prager, D.J. Wilson, G.D. Avery, J.H. Merritt, CA. Garcia, G. Hopen, R. E. Anderson: "Vitreous Fluorophotometry: I d e n t i f i c a t i o n of Sources of V a r i a b i l i t y " , Inves. Ophth. Vis. S c i . , 21, 854 (1981).  17.  J.G. Cunha-Vaz: "Vitreous Fluorophotometry: Techniques, Methodology and C l i n i c a l Applications", handout from Amer. Acad. Ophth. instruction course.  18.  R.C.  19.  J.A. van Best, L. V r i j , J.A. Oosterhuis: "Lens Transmission of Blue-Green Light i n Diabetic Patients as Measured by Autofluorophotometry", Inves. Ophth. Vis. S c i . , 26, 532 (1985).  20.  J.G. Cunha-Vaz, D.M. Maurice: "The Active Transport of Fluorescein by the Retinal Vessels and the Retina", J. Physiol., 191, 467 (1967).  21.  N.P.  22.  B. Krogsaa, H. Lund-Andersen, J. Mehlsen, L. Sestoft, J. Larsen: "The Blood-Retinal Barrier Permeability i n Diabetic Patients", Acta Ophth., 59, 689 (1981).  Zeimer, N.P. Blair, J.G. Cunha-Vaz: "Vitreous Fluorophotometry f o r C l i n i c a l Research: I. Description and Evaluation of a New Fluorophotometer", Arch. Ophth., 101, 1753 (1983).  Zeimer, J.G. Cunha-Vaz, M.E. Johnson: "Studies on the Technique of Vitreous Fluorophotometry", Inves. Ophth. Vis. S c i . , 22, 668 (1982).  Blair, R.C. Zeimer, "Outward Transport Vitreous i n Normal Arch. Ophth., 101,  M.M. Rusin, J.G. Cunha-Vaz: of Fluorescein from the Human Subjects", 1117 (1983).  163  23.  P.S. Chahal, P.J. Chovienczyk, E. M. Kohner: "Measurement of Blood-Retinal Permeability: A Reproducibility Study i n Normal Eyes", Inves. Ophth. Vis. S c i . , 26, 977 (1985).  24.  J.Z. Winkelman, R.J. Zappia, A.J. Gay: "Human Arm to Retina C i r c u l a t i o n Time", Arch. Ophth., 86, 626 (1971).  25.  G.N.  26.  H. Lund-Andersen, B. Krogsaa, J. Larsen, E. Scherfig, L. Sestoft: "Fluorophotometric Evaluation of the Vitreous Body i n the Development of Diabetic Retinopathy", i n Retinal_Diseases, J.G. Cunha-Vaz, ed., Grune & Stratton, Inc., N.Y. (1985), pp. 9-13.  27.  J.G. Cunha-Vaz, R.C. Zeimer, N.P. B l a i r : "Ocular Fuorophotometry", I.S.O.F. Quarterly, 1, 6 (April 1984).  28.  B.P. Conway: "An Analysis of Vitreous Fluorophotometry", i§iiQ§i_Di§§§§§§, J. G. Cunha-Vaz, ed., Grune & Stratton, Inc., N.Y. (1985), pp. 59-66.  Wise, C.T. Dollery, P. Henkind: Ihe_Retinal QilS5=i§ii°Q- Harper & Row, N. Y. (1971), pp. 146-9.  i n  29.  R.C.  Zeimer, N.P. B l a i r , J.G. Cunha-Vaz: "Vitreous Fluorophotometry f o r C l i n i c a l Research: II. Method of Data Acquisition and Processing", Arch. Ophth., 101, 1757 (1983).  30.  B. Krogsaa, H. Fledelius, J. Larsen, H. Lund-Andersen: "Photometric Oculometry. I.: An Analysis of the Optical P r i n c i p l e s i n Slit-Lamp Fluorophotometry", Acta Ophth., 62, 274 (1984).  31.  R.C.  Zeimer, J.G. Cunha-Vaz: "Evaluation and Comparison of Commercial Vitreous Fluorophotometers", Inves. Ophth. Vis. S c i . , 21, 865 (1981).  32.  B.M.  Ishimoto, R.C. Zeimer, J.G. Cunha-Vaz, M.M. Rusin: "Data Analysis and Presentation i n Vitreous Fluorophotometry", I.S.O.F. Quarterly, 2, 1 (Sept. 1985).  33.  J. Larsen. H. Lund-Andersen, B. Krogsaa: "Transient Transport Across the Blood-Retina Barrier", B u l l . Math. B i o l . , 45, 749 (1983).  34.  P.R.  Bevington: Data_Reduction_and_Errgr_An £he_Physical_Sciences, MacGraw-Hill, N.Y. pp. 235-42.  164  (1969),  35.  S-E. Bursell, F. C. Delori, A. Yoshida, J.S. Parker, G.D. Collas, J.W. McMeel: "Vitreous Fluorophotometric Evaluation of Diabetics",' Inves. Ophth. Vis. S c i . , 25, 703 (1984).  36.  G. T.  37.  R.F. Brubaker, R.L. Coakes: "Use of a Xenon Flash Tube as the Excitation Source i n a New Slit-Lamp Fluorophotometer", Amer. J. Ophth., 86, 474 (1978).  38.  E.M.  Kohner, A.R. Alderson: "Vitreous Fluorophotometry", Trans. Ophth. Soc. U.K., 101, 446 (1981).  39.  D.M.  Maurice: "The Use of Fluorescein i n Ophthalmological Research", Inves. Ophth., 6, 464 (1967).  40.  D.W.  H i l l : "The Techniques of Fluorescence Retinography", i n Ih§_yiiii§!D_d§9^§Q2i§_5§Qi§Q§IZ_5yn!E2§iy!D_2D_Ih§ Qsyl§E_Circulatign_in_Health_an J.S. Cant, ed., Kimpton, London, U.K. (1969), p. 53.  41.  F. Delori, I. Ben-Sira, C. Trempe: "Fluorescein Angiography with an Optimized F i l t e r Combination", Am. J. Ophth., 82, 559 (1976).  42.  N.P.  43.  P. Horowitz, W. H i l l : Ihe_Art_of.Electronics, Cambridge U. Press, N.Y. (1980).  44.  T. Hogan, M. Iannamico: QSBORNE_l_Technical_Manual, Osborne Computer Corp., CA. (1982), Part No. 2F00153-01.  45.  T. Hogan, M. Iannamico: QSBORNE_l_User^s_Reference_Guide, Osborne Computer Corp., CA. (1981).  46.  A.R.  M i l l e r : §080/Z280_Assembly_Language2_Te Imprgved_Prggrarnming, Wiley, N. Y. (1981).  47.  R.C.  Zeimer, N.P. B l a i r , J.G. Cunha-Vaz: "Pharmacokinetic Interpretation of Vitreous Fluorophotometry", Inves. Ophth. Vis. S c i . , 24, 1374 (1983).  Feke, R. Zuckerman, G.J. Green, J.J. Weiter: "Response of Human Retinal Blood Flow to Light and Dark", Inves. Ophth. Vis. S c i . , 24, 136 (1983).  B l a i r , M.M. Rusin, E. Shakin: "The Effect of pH on the Transfer of Fluorescein across the BloodRetinal Barrier", Inves. Ophth. Vis. S c i . , 26, 1133 (1985).  165  48.  G.F. Carrier, C.E. Pearson: P§rtial_Differential_Eguatigns Jhfgr^_and_Technigue, Academic Press, Inc. , N.Y. (1976), pp.257-81.  49.  W.A.  Ray, D.M. O'Day: " S t a t i s t i c a l Analysis of Multi-Eye Data i n Ophthalmic Research", Inves. Ophth. Vis. S c i . , 26, 1186 (1985).  50.  S.N.  Markowitz, J.D. Morin: "The Ratio of Lens Thickness to Axial Length f o r Biometric Standardization i n Angle-Closure Glaucoma", Amer. J. Ophth., 99, 400 (1985).  51.  G.W.  Snedecor, W.G. Cochran: Statistigal_Methods, 6th ed., The Iowa State U. Press, IA. (1967),  52.  R.E. Walpole: Intrgductign_to_Statistics, 3rd ed., MacMillan, N.Y. (1982).  53.  Y. Ogura, Y. Tsukahara, I. Saito, T. Kondo: "Estimation of the Permeability of the Blood-Retinal Barrier in Normal Individuals", Inves. Ophth. Vis. S c i . , 26, 969 (1985).  54.  H. Lund-Andersen, B. Krogsaa, M. l a Cour, J. Larsen: "Quantitative Vitreous Fluorophotometry Applying a Mathematical Model of the Eye", Inves. Ophth. Vis. S c i . , 26, 698 (1985).  166  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics

Country Views Downloads
China 84 13
United States 16 0
Japan 7 0
Brazil 3 0
Unknown 2 1
Indonesia 2 0
Ukraine 1 0
Singapore 1 0
Poland 1 0
Mexico 1 0
City Views Downloads
Hangzhou 28 0
Unknown 16 13
Beijing 8 5
Guangzhou 8 0
Shenzhen 6 8
Ashburn 5 0
Jinan 4 0
Shanghai 4 0
Lawrenceville 3 0
Harbin 3 0
Clarks Summit 3 0
Guiyang 3 0
Kunming 3 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}
Download Stats



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items