Towards a high resolution retinal implant

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content TOWARDS A HIGH RESOLUTION RETINAL IMPLANT

By

Alejandra Gonzalez Calle

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOMEDICAL ENGINEERING)

May 2018

Copyright 2018 Alejandra Gonzalez Calle
ii

Epigraph

Knowledge is the eye of desire and can become the pilot of the soul.

-Will Durant

iii
Acknowledgments

Firstly, I would like to express my sincere gratitude to my advisor Prof.
James D. Weiland for the continuous support during my Ph.D study and during
these nine years, for his patience, motivation, and immense knowledge. His
guidance helped me in all the time of research and writing of this thesis. I am who I
am, and I am where I am because nine years ago Dr. Weiland decided to give me an
opportunity and accepted me in his laboratory as a volunteer and I will always be
thankful for that. I could not have imagined having a better advisor and mentor
during these years.

Besides my advisor, I would like to thank the rest of my thesis committee:
Prof. D’Argenio, Prof. Humayun, and Dr. Kashani, for their insightful comments and
encouragement, and advice they provided me over the years.

I thank my fellow labmates in for the stimulating discussions, and all their
help during my research. I want to specially thank Leanne Chan for sharing her
knowledge with me when I first joined the lab and Navya Davuluri, Aditi Ray, Artin
Petrossians, Karthik Murali, Kiran Nimmagadda, Laura Liu, Sabina Morales, Jose
Daniel Barbosa, Bruno Diniz, Yi Zhang, Rodrigo Brant, Juan Carlos Martinez and
also Mort Arditti, Fernando Gallardo and Lina Flores for all their technical support
and patience. Also I want to thank all my friends from school Ivan Trujillo, Enrique
Arguelles, Luisa Obregon, Gene Yu, Uldric and John Sunwoo for all the moral
iv
support. As I always say, you guys make it look so easy that it makes me push
myself to try to keep up with you.

I would also like to thank my collaborators at the University of Michigan that
helped me making this last year go smooth, for all the hard work and waking up
early and staying up late. Without all your help this last semester would have been
very difficult. Special Thanks to Yooree Chung, Swetha Jeganathan, Prabha
Narayanaswamy and Gail Rising. I would also like to thank the employees of Second
Sight Medical Products, Inc. including Weston Pack, Jessy Dorn, Vara Wuyyuru and
Uday Patel.

I am very thankful for my family: my parents Helena Calle and Gerardo
Gonzalez, my brother Gerardo Gonzalez and my soul sister Luz Marina Montealegre
for supporting me during my PhD carrer and throughtout my life, for giving me a
good foundation to be the person I am today, for all the advice and encouragement
during the good and tough times.

Last but not least I would like to thank Leonardo Nava and Akira for being
my family and my support system, for all the unconditional love, for all the support
and for all the patience.
v
Table of Content
EPIGRAPH .......................................................................................................... II
ACKNOWLEDGMENTS ................................................................................... III
TABLE OF CONTENT ........................................................................................ V
LIST OF TABLES ............................................................................................ VIII
LIST OF FIGURES ............................................................................................. IX
ABSTRACT ........................................................................................................... 1

CHAPTER 1

INTRODUCTION
1.1. EYE OVERVIEW ............................................................................................................................................ 4
1.2 RETINA - ANATOMY ..................................................................................................................................... 5
1.2.1 Photoreceptor layer ............................................................................................................................. 6
1.2.2 Bipolar layer ............................................................................................................................................ 7
1.2.3 Ganglion cell layer ................................................................................................................................ 8
1.2.4 Muller Cells ............................................................................................................................................... 9
1.3 RETINA –VISUAL PATHWAY ........................................................................................................................ 9
1.4 RETINA DISEASES ...................................................................................................................................... 11
1.4.1 Aged- Related Macular Degeneration (AMD) ........................................................................ 11
1.4.2 Retinitis Pigmentosa (RP) .............................................................................................................. 12
1.5 TREATMENTS ............................................................................................................................................. 13
1.5.1 Photodynamic Therapy ................................................................................................................... 14
1.5.2 Anti- VEGF ............................................................................................................................................. 14
1.5.3 Gene Therapy ....................................................................................................................................... 14
1.5.4 RPE Transplantation ........................................................................................................................ 15
1.5.5 Stem-Cell-Based Therapies ............................................................................................................ 15
1.6 VISUAL PROSTHESES ................................................................................................................................. 15
1.6.1 History ..................................................................................................................................................... 16
1.7 RETINAL PROSTHESES .............................................................................................................................. 18
1.7.1 Epiretinal prostheses ........................................................................................................................ 19
1.7.1.1 Second Sight Medical Products ............................................................................................................................. 20
1.7.1.2 Epi-Ret ............................................................................................................................................................................. 24
1.7.1.3 Intelligent Medical Implants .................................................................................................................................. 25
1.7.2 Subretinal Prostheses ....................................................................................................................... 27
1.7.2.1 Optobionics .................................................................................................................................................................... 28
1.7.2.2 Retina Implant AG ....................................................................................................................................................... 30
1.7.2.3 Boston Retinal Implant Project ............................................................................................................................. 32
1.7.2.4 Stanford University .................................................................................................................................................... 34
1.7.3 Suprachoroidal Prostheses ............................................................................................................ 35
1.7.3.1 Bionic Vision Australia (BVA) ................................................................................................................................ 36
1.7.3.2 Department of Visual Science – Osaka University ........................................................................................ 38
vi
1.8 PROBLEMS WITH RETINAL PROSTHESIS ................................................................................................ 39
1.8.1 Electrode-Retina Distance .............................................................................................................. 40
1.8.2 Electrode size ....................................................................................................................................... 40
1.8.3 Axonal stimulation ............................................................................................................................ 41
1.8.4 Stimulation selectivity ...................................................................................................................... 41

CHAPTER 2

DISINTEGRIN-INTEGRIN NOVEL ATTACHMENT TO HELP WITH
ELECTRODE-RETINA DISTANCE ............................................................... 43
2.1 BACKGROUND ............................................................................................................................................. 43
2.1.1 Importance of retinal distance ..................................................................................................... 43
2.1.2 Epiretinal prosthesis – Tack .......................................................................................................... 45
2.1.3 Integrins ................................................................................................................................................. 46
2.1.4 Disintegrins ........................................................................................................................................... 49
2.1.5 Silicone .................................................................................................................................................... 50
2.2 SPECIFIC AIM .............................................................................................................................................. 52
2.3 MATERIAL AND METHODS ....................................................................................................................... 53
2.3.1 Silicone- Disintegrin Process ......................................................................................................... 53
2.3.2 Silicone Sterilization ......................................................................................................................... 55
2.3.3 In-vitro Experiments ......................................................................................................................... 56
2.3.4 Surgical Procedure – In vivo Experiments .............................................................................. 56
2.4 RESULTS ...................................................................................................................................................... 58
2.4.1 Silicone - Disintegrin process ........................................................................................................ 58
2.4.2 In-vitro Experiments ......................................................................................................................... 62
2.4.3 Sterilization Protocol ....................................................................................................................... 63
2.4.4 In-vivo Experiments .......................................................................................................................... 65
2.5 DISCUSSION AND CONCLUSION ............................................................................................................... 70
CHAPTER 3

HIGH CHARGE DENSITY STIMULATION SAFETY ................................. 74
3.1 BACKGROUND ............................................................................................................................................. 74
3.1.1 Neural stimulation ............................................................................................................................ 74
3.1.2 The electrode- Tissue interface .................................................................................................... 75
3.1.3 Equivalent circuit model ................................................................................................................. 77
3.1.4 Stimulation Characteristics ........................................................................................................... 78
3.1.4.1 Cathodic and anodic stimulation .......................................................................................................................... 78
3.1.4.2 Monopolar and bipolar stimulation .................................................................................................................... 79
3.1.4.3 Voltage controlled vs. Current Controlled stimulation .............................................................................. 80
3.1.5 Stimulation Electrode ...................................................................................................................... 81
3.1.6 Safety stimulation .............................................................................................................................. 83
3.1.6.1 Causes of electrical damage .................................................................................................................................... 83
3.3 SPECIFIC AIM .............................................................................................................................................. 90
3.4 MATERIALS AND METHODS ..................................................................................................................... 92
3.4.1 Animal Model ....................................................................................................................................... 92
3.4.2 Anesthesia .............................................................................................................................................. 93
3.4.2.1 University of Southern California ........................................................................................................................ 93
3.4.3 Surgical Procedures .......................................................................................................................... 93
vii
3.4.3.1 Non-Survival Experiments- University of Southern California ............................................................... 93
3.4.4 Stimulation Electrode Insertion .................................................................................................. 96
3.4.5 Stimulation Electrode ...................................................................................................................... 98
3.4.6 Experimental Setup ........................................................................................................................... 99
3.4.7 Experimental Groups ..................................................................................................................... 100
3.4.8 Data Analysis .................................................................................................................................... 102
3.4.8.1 Imaging analysis ....................................................................................................................................................... 102
3.4.8.2 Statistical Analysis ................................................................................................................................................... 108
3.5 RESULTS .................................................................................................................................................... 108
3.5.1 Non-Survival ...................................................................................................................................... 108
3.5.1.1 Electrode-Retina Distance .................................................................................................................................... 108
3.5.1.2 Stimulation-Retinal Thickness ........................................................................................................................... 111
3.5.2 Survival ................................................................................................................................................ 124
3.6 CONCLUSION ............................................................................................................................................. 126

CHAPTER 4

TEST STIMULUS PROTOCOLS IN RETINAL-PROSTHESIS PATIENTS
........................................................................................................................... 129
3.1 BACKGROUND ........................................................................................................................................... 129
3.1.1 Retina response to electrical stimulation ............................................................................. 129
3.2 PREVIOUS RESEARCH .............................................................................................................................. 131
3.3 SPECIFIC AIM ............................................................................................................................................ 137
3.4 METHODS .................................................................................................................................................. 138
3.4.1 System .................................................................................................................................................. 138
3.4.2 Subjects ................................................................................................................................................ 139
3.4.2.1 Subject 1 ....................................................................................................................................................................... 139
3.4.2.2 Subject 2 ....................................................................................................................................................................... 139
3.4.2.3 Subject 3 ....................................................................................................................................................................... 140
3.4.2.4 Subject 4 ....................................................................................................................................................................... 140
3.4.2.5 Subject 5 ....................................................................................................................................................................... 141
3.4.3 Psychophysical Tests ...................................................................................................................... 141
3.4.3.1 Tactile – Drawing Task .......................................................................................................................................... 141
3.4.3.2 Stimulation – Drawing Task ................................................................................................................................ 143
3.4.3.3 Retinal stimulation parameters ......................................................................................................................... 145
3.4.4 Analysis ................................................................................................................................................ 145
3.5 RESULTS .................................................................................................................................................... 146
3.5.1 Subject 1 .............................................................................................................................................. 146
3.5.2 Subject 2 .............................................................................................................................................. 150
3.5.3 Subject 3 .............................................................................................................................................. 154
3.5.4 Subject 4 .............................................................................................................................................. 159
3.5.5 Subject 5 .............................................................................................................................................. 164
3.6 CONCLUSIONS ........................................................................................................................................... 166

CHAPTER 5

CONCLUSIONS AND FUTURE WORK ...................................................... 169
REFERENCES .................................................................................................. 173

viii

List of Tables
TABLE 1. AN ATL PROMASTER EXCIMER LASER WAS USED DURING ALL THE EXPERIMENTS PERFORMED. SHOWN IN THIS
TABLE ARE SETTINGS THAT SHOWED A CONSISTENT DEBRIS FIELD AREA ON THE SILICONE SURFACE, THUS WERE
USED DURING ALL THE EXPERIMENTS REPORTED HERE. ..................................................................................................... 54
TABLE 2. SEVEN DIFFERENT TYPES OF SILICONES WERE LASED AND TREATED WITH DISINTEGRINS TO EVALUATE THE
DEGREE OF DISINTEGRIN BINDING. SAMPLES WERE EVALUATED FOR GROWTH OF OVCAR3 BY VISUAL INSPECTION.
INCREASE IN FLUORESCENCE MEANS VCN-SILICONE SUPPORTS CELL GROWTH WHICH IT IS PROPORTIONAL TO
DISINTEGRIN BINDING. ............................................................................................................................................................. 62
TABLE 3. IN-VITRO EXPERIMENTS IN CADAVERIC PIG EYES WERE PERFORMED TO STUDY DISINTEGRIN- RETINA
ATTACHMENT. FOUR DIFFERENT TYPES OF SILICONES AND TWO DIFFERENT TYPES OF DISINTEGRINS WERE USED
DURING THESE EXPERIMENTS. ATTACHMENT OF SILICONE SAMPLES WITH TWO DIFFERENT DISINTEGRINS. ----
INDICATES NO OBSERVABLE BINDING; + THROUGH +++ INDICATE INCREASING LEVELS OF BINDING. ........................ 63
TABLE 4. DIFFERENT STERILIZATION PROTOCOLS WERE TESTED TO ENSURE VIABILITY OF THE DISINTEGRIN AFTER
STERILIZATION. GROUP 1 WENT THROUGH THE LASING PROCESS WITHOUT ANY STERILIZATION METHOD. GROUG 2
WAS STERILIZED BEFORE THE LASING PROCESS. GROUP 3 WAS STERILIZED AFTER THE LASING PROCESS AND GROUP
4 IS A CONTROL GROUP, SILICONES WERE NO LASED AND PLACED IN PBS. ---- INDICATES NO OBSERVABLE BINDING;
+ THROUGH +++ INDICATE INCREASING LEVELS OF BINDING. ........................................................................................... 65
TABLE 5. PARAMETERS USED DURING STIMULATION. EXPERIMENTAL GROUPS WERE BASED ON DIFFERENT
COMBINATIONS OF CHARGE DENSITIES, STIMULUS FREQUENCY, PULSE WIDTHS, AND DURATION OF PULSE TRAIN
STIMULATIONS. ........................................................................................................................................................................ 100
TABLE 6. PARAMETERS USED DURING STIMULATION AND SURVIVAL EXPERIMENTS. EXPERIMENTAL GROUPS WERE BASED
ON DIFFERENT COMBINATIONS OF CHARGE DENSITIES, STIMULUS FREQUENCY, PULSE WIDTHS, AND DURATION OF
PULSE TRAIN STIMULATIONS. THE GREEN REPRESENTS THE GROUPS THAT SHOWED NO CHANGES IN RETINAL
THICKNESS DURING NON-SURVIVAL EXPERIMENTS, AND THE RED ARE THE GROUPS THAT SHOWED CHANGES IN
RETINAL THICKNESS DURING NON-SURVIVAL EXPERIMENTS. .......................................................................................... 102

ix

List of Figures
FIGURE 1. CROSS SECTION OF THE HUMAN EYE. WHEN LIGHT STRIKES THE EYE, IT PASSES THROUGH THE CORNEA AND
THE LENS TO FORM AN IMAGE TO THE RETINA. VISUAL INFORMATION IS TRANSMITTED BY THE OPTIC NERVE TO THE
HIGHER VISUAL CENTERS OF THE BRAIN ................................................................................................................................... 5
FIGURE 2. RETINA STRUCTURE. SCHEMATIC OF THE DIFFERENT CELL TYPES AND RETINAL LAYERS ENCOUNTERED IN THE
RETINA. THE RETINA CONVERTS LIGHT INTO ELECTRICAL SIGNALS THAT ARE CARRIED BY THE OPTIC NERVE TO THE
BRAIN. PROCESS STARTS AT THE PHOTORECEPTORS LAYER, AND THEN PASSED THROUGH EACH LAYER OF THE
RETINA TO THE GANGLION CELLS. GANGLION CELLS TRANSMIT THE ACTION POTENTIALS TO THE CORTEX FOR
FURTHER PROCESSING. ................................................................................................................................................................ 6
FIGURE 3. REPRESENTATION OF PATIENT’S WITH NORMAL VISION (LEFT) AND PATIENT’S WITH AMD (RIGHT). AMD
CAN VARY FROM SLIGHT VISUAL DISTORTIONS TO COMPLETE LOSS OF CENTRAL VISION. .............................................. 12
FIGURE 4. REPRESENTATION OF PATIENT’S WITH NORMAL VISION (LEFT) AND PATIENT’S WITH RETINITIS PIGMENTOSA
(RIGHT). PATIENTS WITH RETINITIS PIGMENTOSA HAVE LOSS OF NIGHT VISION, CONSTRICTED VISUAL FIELDS, AND
COMPLETE BLINDNESS IN ITS LATE STAGES. ......................................................................................................................... 13
FIGURE 5. LEFT IMAGE. BRINDLEY’S CORTICAL PROSTHESIS SHOWN PRIOR TO IMPLANTATION. RIGHT IMAGE. X-RAY OF
THE CORTICAL PROSTHESIS POST IMPLANTATION. THE ARROW SHOWS THE RECEIVER ARRAY, AND THE ARROWHEAD
SHOWS THE ELECTRODE ARRAY. ............................................................................................................................................. 17
FIGURE 6. ILLUSTRATION SHOWING THE EYE STRUCTURE, THE STRATIFIED STRUCTURE OF THE RETINA, AND THE
PLACEMENT FOR EPIRETINAL, SUBRETINAL, AND SUPRACHOROIDAL PROSTHESES. ....................................................... 19
FIGURE 7. ARGUS EPIRETINAL PROSTHESES: (A) EXTERNAL PARTS OF THE ARGUS SYSTEM. (B) EXTRAOCULAR AND
INTRAOCULAR PART OF THE ARGUS II SYSTEM. (C) FUNDUS IMAGE OF A PATIENT IMPLANTED WITH THE ARGUS I
SYSTEM. ARGUS I IMPLANT CONTAINS 16 ELECTRODES. (D) FUNDUS IMAGE OF A PATIENT IMPLANTED WITH THE
ARGUS II SYSTEM. ARGUS II IMPLANT CONTAINS 60 ELECTRODES. .................................................................................. 22
FIGURE 8. EPIRETINAL PROSTHESIS. (A) SCHEMATIC OF THE INTELLIGENT MEDICAL IMPLANT (IMI) SYSTEM. (B-C)
IMI PROTOTYPE. ........................................................................................................................................................................ 27
FIGURE 9. SUBRETINAL PROSTHESES. (A) SCHEMATIC OF THE BOSTON RETINAL IMPLANT SHOWING THE SECONDARY
COIL SURROUNDING THE CORNEA. (B-C) RETINAL IMPLANT FROM RETINAL IMPLANT AG CALLED ALPHA IMS. IT
CONSISTS OF THE VISION CHIP (MULTIPHOTODIODES ARRAY) ON A POLYIMIDE FOIL (BOTH PLACED SUBRETINALLY),
A POWER SUPPLY CABLE CONNECTING THE MICROCHIP WITH THE RECEIVER COIL IN A CERAMIC HOUSING, AND THE
REFERENCE ELECTRODE PLACED SUBDERMALLY AT THE TEMPLE AND RETROAURICULAR REGION. ............................ 32
FIGURE 10. SUPRACHOROIDAL PROSTHESES. (A) SUPRACHOROIDAL IMPLANT. THIS IMPLANT CONTAINS 49
ELECTRODES. (B) BIONIC VISION AUSTRALIA (BVA) IMPLANT CONSISTS OF ONE REMOTE RETURN, TWO OTHER
RETURN ELECTRODES ON THE SUPRACHOROIDAL ARRAY, AND A CHIP OF 33 PLATINUM STIMULATING ELECTRODES.
...................................................................................................................................................................................................... 37
FIGURE 11. THE EFFECT OF ELECTRODE-RETINA DISTANCE. STUDY PERFORMED IN PATIENTS IMPLANTED WITH THE
ARGUS II SYSTEM. THE MEAN ELECTRODE THRESHOLD ACROSS ALL SUBJECTS WAS 206 ±6.3 ΜA (N=703
ELECTRODES) AND THE MEAN ELECTRODE-RETINA DISTANCE WAS 179.6 ±6.5 ΜM (N=1013 ELECTRODES).
[40] ............................................................................................................................................................................................. 44
FIGURE 12. RETINAL TACKING. LEFT IMAGE. ILLUSTRATION OF AN ARGUS I ARRAY PLACED EPIRETINALLY, SHOWING
THE INSTRUMENT AND TECHNIQUE USE TO TACK THE ARRAY TO THE RETINA. RIGHT IMAGE. FUNDUS IMAGE OF A
PATIENT IMPLANTED WITH THE ARGUS II ARRAY WHERE THE TACK IS EASILY VISIBLE ON THE TOP OF THE ARRAY.
...................................................................................................................................................................................................... 46
x
FIGURE 13. ILLUSTRATION OF INTEGRIN-DISINTEGRIN INTERACTION. INTEGRINS ARE A FAMILY OF TRANSMEMBRANE
RECEPTOR PROTEINS, THAT BIND TO COMPONENTS OF THE EXTRACELLULAR MATRIX. DISINTEGRINS CONTAIN A
SECONDARY STRUCTURE KNOWN AS “RGD ADHESIVE LOOP”, WHICH PLAYS A VERY IMPORTANT ROLE IN THE
INTERACTION WITH SPECIFIC INTEGRIN RECEPTORS. .......................................................................................................... 48
FIGURE 14. SILICONE STRUCTURE. SILICONES CONSIST OF AN INORGANIC SILICON (SI) – OXYGEN (O) BACKBONE CHAIN
WITH ORGANIC SIDE GROUPS (R) ATTACHED TO THE SILICON ATOMS. THE ORGANIC GROUPS ARE USUALLY METHYL
GROUPS (EXAMPLE: CH3). ...................................................................................................................................................... 51
FIGURE 15. RABBIT RETINA INTEGRINS. (A) PANEL ON LEFT DEMONSTRATES Α5 INTEGRIN (YELLOW ARROW, RED
FLUORESCENCE) IN THE ILM. (B) PANEL IN THE MIDDLE SHOWS ABSENCE OF Β3. (C) PANEL ON RIGHT
DEMONSTRATES Β5 INTEGRIN PRESENCE (WHITE ARROW, GREEN FLUORESCENCE) IN THE ILM OF THE RABBIT
RETINA. BLUE FLUORESCENCE IS DAPI COUNTER STAINING OF THE NUCLEI OF THE NEURONS. GREY IMAGES ARE
PHOTOMICROGRAPHS. LOWER RIGHT IMAGE OF EACH PANEL IS OVERLAY OF ALL THREE IMAGES. IMAGE TAKEN FROM
[104]. ......................................................................................................................................................................................... 52
FIGURE 16. DIFFERENT TYPES OF SILICONES WERE LASED TO EVALUATE IF AN ACTIVE AREA COULD BE CREATED. SEM
IMAGES WERE TAKEN TO EACH TYPE OF SILICONE AFTER LASING PROCESS. EVALUATION CONSISTED ON VISUALLY
ASSESSING EACH SAMPLE FOR PHYSICAL ABLATION/DEBRIS FIELD AREA ON THE SURFACE, THE AMOUNT OF
MATERIAL REMOVED, DEFINITION OF EDGES, EVENNESS, DEPTH, AND SYMMETRY OF THE LASED AREA. THE
SILICONES SHOWN IN THIS IMAGE ARE A) MED 4286 B) WL5150 C) MED4830 D) MED4840 E) MED4850
F) MED4860. MED 4850 AND MED 4860 HAVE EDGES THAT ARE CLEAN AND DEFINED, UNIFORM ABLATION
AREA, AND A LARGE DEBRIS FIELD. ......................................................................................................................................... 59
FIGURE 17. IMAGES TAKEN TO EVALUATE DISINTEGRIN BINDING AND CELL GROWTH ON SILICONES AFTER LASING
PROCESS AND DISINTEGRIN EXPOSURE: A) MED 4860 SIX HOURS AFTER PLATING CELLS AND B) MED 4860 48
HOURS AFTER PLATING CELLS SHOWS CELL GROWTH BY LIGHT MICROSCOPY ON THE LASED AREA BUT NOT ON THE
UNLASED AREAS THAT WERE ALSO EXPOSED TO DISINTEGRIN. C) MED4830 EXPOSED TO OVCAR-3, D)
MED4840, E) MED4850, AND F) MED4860. C-F WERE EXPOSED TO OVCAR -3 AND CELL GROWTH IS
EVALUATED BY FLUORESCENCE. MED 4830 TO MED 4860 SUPPORT CELL GROWTH BUT MED 4860 IS SUPERIOR
IN SUPPORTING BIOLOGICAL FUNCTION OF THE DISINTEGRINS (INCREASE IN FLUORESCENCE OBSERVED DUE TO
INCREASED OVCAR3 CELL GROWTH). .................................................................................................................................. 61
FIGURE 18. IMAGES WERE TAKEN TO EVALUATE DISINTEGRIN BINDING IN SILICONES AFTER LASING PROCESS, RETINA-
SILICONE ATTACHMENT AND FEASIBILITY OF STERILIZATION METHOD. DURING EACH TIME POINT, THREE IMAGES
WERE TAKEN. COLUMN A SHOWS A FUNDUS IMAGE OF THE RETINA WITH THE SILICONE SAMPLE ATTACHED TO IT.
COLUMN B SHOWS AN OCT IMAGE. THIS IS A CROSS SECTION OF THE RETINA WHERE THE SAMPLE IS ATTACHED AND
COLUMN C SHOWS A FLUORESCEIN ANGIOGRAPHY IMAGE TO EVALUATE RETINAL CIRULATION. IMAGES WERE TAKEN
AT 5 DIFFERENT TIME PERIODS: ROW 1) ONE WEEK AFTER IMPLANTATION, ROW 2) TWO WEEKS AFTER
IMPLANTATION, ROW 3) 5 WEEKS AFTER IMPLANTATION, ROW 4) 2 MONTHS AFTER IMPLANTATION AND ROW 5)
3 MONTHS AFTER IMPLANTATION. ......................................................................................................................................... 68
FIGURE 19. IN-VIVO EXPERIMENT. ATTACHMENT OF STERILE SILICONE SAMPLES WITH DISINTEGRINS. – INDICATES NO
OBSERVABLE ATTACHMENT OR INFECTION; + INDICATES ATTACHMENT OR INFECTION; X INDICATES DATA POINT
WAS NOT ACQUIRED. AT, ATTACHMENT, IN, INFECTION. 1WEEK (1W), 2WEEKS (2W), 5 WEEKS (5W), 2MONTHS
(2M), 3MONTHS (3M) INDICATES THE DATA POINTS WHERE RABBITS WERE EVALUATED. RABBIT 1(R1) AND
RABBIT 2(R2) SHOWED ATTACHMENT DURING THE PERIOD THEY WERE OBSERVED AND NO SIGNS OF INFECTION
DURING THE SAME PERIOD. RABBIT 3(R3) AND RABBIT (R4) SHOWED NEITHER ATTACHMENT NOR SIGN OF
INFECTION DURING THE PERIOD THEY WERE MAINTAINED. ............................................................................................... 69
FIGURE 20. HISTOLOGY IMAGES TAKEN FROM RABBIT 2. (A) ARROWS SHOW AREA WHERE SAMPLE WAS PLACED. (B)
ZOOMED IMAGE OF AREA BETWEEN ARROWS IN IMAGE A. NO SIGNIFICANT DAMAGE IS RECORDED IN THESE IMAGES.
IMAGE B SHOWS SOME RED BLOOD CELLS ON THE RETINAL SURFACE (ARROWS POINT TO THE RED BLOOD CELLS). 70
FIGURE 21. NEURON SCHEMATIC. NEURONS ARE EXCITABLE CELLS THAT FUNCTION TO PROCESS AND TRANSMIT
INFORMATION. INFORMATION TRAVELS FROM NEURON TO NEURON EITHER BY ELECTRICAL OR CHEMICAL SIGNALS.
...................................................................................................................................................................................................... 75
FIGURE 22. (A) DEVICE-TISSUE INTERFACE. (B) A DYNAMIC MODEL AT THE DEVICE-TISSUE INTERFACE WHEN A
CURRENT IS APPLIED AT THE ELECTRODE. ............................................................................................................................ 77
FIGURE 23. MONOPOLAR AND BIPOLAR STIMULATION CONFIGURATION. CURRENT IS DISTRIBUTED EVENLY IN ALL
DIRECTIONS WITH A MONOPOLAR CONFIGURATION. CURRENT IS CONCENTRATED BETWEEN THE TWO ELECTRODES
LEADING TO A LOCALIZED STIMULATION WITH A BIPOLAR CONFIGURATION. .................................................................. 80
xi
FIGURE 24. THE HATCHED AREA REPRESENTS THE REGION OF CHARGE AND CHARGE DENSITY WHERE NEURAL DAMAGE
IS OBSERVED. THE SOLID LINES REPRESENT THE PROPOSED MODEL WITH PARAMETERS K=1.0, 1.5 AND 2.0. ....... 87
FIGURE 25. IMAGES OF A RAT’S RETINA WITH HEMATOXYLIN STAINING. H&E STAINING SHOWED NO MORPHOLOGICAL
CHANGES TO ANY OF THE EXPERIMENTAL CONDITIONS EVALUATED IN THIS PAPER. FIGURE B IS A CONTROL RETINA
AND FIGURE C IS A RETINA STIMULATED AT A HIGH FREQUENCY. NO LAYER DISORGANIZATION IS OBSERVED IN THE
MAGNIFICATION INSETS. ........................................................................................................................................................... 88
FIGURE 26. OCT IMAGES AT DIFFERENT TIME POINTS DURING STIMULATION. FOUR DIFFERENT CHARGE DENSITIES
WERE EVALUATED. NUMBER ON TOP OF EACH QUADRANT (4 QUADRANTS) REPRESENTS THE CHARGE DENSITY
USED DURING THAT SPECIFIC EXPERIMENT. EACH QUADRANT IS COMPOSED OF THREE OCT IMAGES. FROM TOP TO
BOTTOM, FIRST IS AN OCT IMAGE ZERO MINUTES INTO THE STIMULATION, MIDDLE IMAGE IS AN OCT IMAGE RIGHT
AFTER STIMULATION WAS FINISHED AND LAST IMAGE IS AN OCT IMAGE FIFTEEN MINUTES AFTER STIMULATION
WAS OVER. .................................................................................................................................................................................. 90
FIGURE 27. CHARGE DENSITY FOR DIFFERENT ELECTRODE SIZES. INJECTING THE SAME AMOUNT OF CHARGE THROUGH
THREE DIFFERENT SIZE ELECTRODES. THE SMALLEST ELECTRODE REQUIRES THE HIGHEST CHARGE DENSITY. ....... 91
FIGURE 28. LEFT IMAGE SHOWS THE SETUP USED DURING THE NON-SURVIVAL EXPERIMENTS PERFORMED IN THE
UNIVERSITY OF SOUTHERN CALIFORNIA. RIGHT IMAGE SHOWS THE SETUP USED DURING SURVIVAL EXPERIMENTS
PERFORMED IN THE UNIVERSITY OF MICHIGAN. ................................................................................................................. 96
FIGURE 29. DIAGRAM OF THE EXPERIMENT SETUP. THE ELECTRODE WAS PLACED THROUGH THE SCLERA AND HELD IN
PLACE WITH A MICROMANIPULATOR. RETURN ELECTRODE WAS PLACED IN THE RABBIT’S HEAD. A SPECTRAL
DOMAIN OPTICAL COHERENCE TOMOGRAPHY (OCT) SYSTEM WAS PLACED IN FRONT OF THE RABBIT’S LEFT EYE AND
SCANNED THE RETINA SURFACE WHILE THE RETINA WAS BEING STIMULATED. ............................................................. 97
FIGURE 30. STIMULATION ELECTRODE SCHEMATIC. ELECTRODE WAS CUSTOM MADE BY FHC, INC. .................................. 98
FIGURE 31. DIAGRAM OF EXPERIMENTAL SETUP. A MCS SYSTEM DELIVERS THE STIMULUS PULSES TO THE RETINA.
WORKING ELECTRODE GOES IN THE VITREOUS CAVITY AND RETURN ELECTRODE IN THE RABBIT’S HEAD. CURRENT
AND VOLTAGE WAVEFORMS ARE MONITORED WITH AN OSCILLOSCOPE. ........................................................................... 99
FIGURE 32. FUNDUS IMAGE OF THE RABBIT’S EYE. ELECTRODE WAS PLACED ON THE INFERIOR TEMPORAL REGION OF
THE EYE, CLOSE TO THE VISUAL STREAK. LEFT IMAGE IS ACQUIRED WITH THE SPECTRALLIS OCT SYSTEM. RIGHT
IMAGE IS ACQUIRED WITH THE BIOPTIGEN OCT SYSTEM ................................................................................................. 103
FIGURE 33. TOP LEFT IMAGE. FUNDUS IMAGE OF THE RABBIT’S EYE. THE X SCAN LINE WAS ADJUSTED TO BE ON THE TIP
OF THE ELECTRODE SIMULTANEOUSLY WHILE THE ELECTRODE WAS ADVANCED TOWARD THE RETINA; WHICH
HELPED FOR PLACEMENT OF THE ELECTRODE CLOSE TO THE RETINA. TOP RIGHT IMAGE. OCT IMAGE. ELECTRODE
TIP WAS PLACED CLOSE TO THE EPIRETINAL SURFACE. ELECTRODE WAS ADVANCED TOWARDS THE RETINA UNTIL
THE TIP OF THE ELECTRODE COULD BE SEEN ON THE OCT SCAN. DISTANCE BETWEEN THE ELECTRODE AND THE
RETINA WAS MEASURED USING THE CALIPERS FUNCTION OF THE OCT SOFTWARE (SHOWN IN RED). TOP IMAGES
WERE ACQUIRED WITH THE SPECTRALLIS OCT SYSTEM. BOTTOM IMAGES WERE ACQUIRED WITH THE BIOPTIGEN
OCT SYSTEM ............................................................................................................................................................................ 104
FIGURE 34. RETINAL THICKNESS MEASUREMENT. TOP LEFT IMAGE IS A FUNDUS IMAGE OF THE RABBIT’S EYE WITH THE
X SCAN LINE SHOWING THE SECTION OF THE RETINA SHOWN IN THE RIGHT IMAGE. TWO RED LINES IN THE TOP
RIGHT IMAGE ARE USED TO MEASURE THE RETINAL THICKNESS. SOFTWARE MEASURE DISTANCE BETWEEN THE TWO
RED LINES AS SHOWN WITH THE PERPENDICULAR GREEN LINE IN THE TOP RIGHT IMAGE AND IT IS REPORTED AS
SHOWN IN THE MIDDLE IMAGE. TOP AND MIDDLE IMAGES WERE ACQUIRED WITH THE SPECTRALLIS OCT SYSTEM.
BOTTOM IMAGES WERE ACQUIRED WITH THE BIOPTIGEN OCT SYSTEM. BOTTOM LEFT IMAGE SHOWS THE RETINAL
THICKNESS BEING MEASURED WITH MULTIPLE CALIPERS IDENTIFIED WITH DIFFERENT COLORS EACH. BOTTOM
RIGHT IMAGE SHOWS A FUNDUS IMAGE WHERE EACH COLOR THAT REPRESENTS THE CALIPER IS SHOWN TO
REPRESENT THE SECTION OF THE RETINA THAT IS BEING MEASURED. ........................................................................... 106
FIGURE 35. FLUORESCENT ANGIOGRAPHY IMAGE. IMAGE OF A RABBIT’S RETINA AFTER STIMULATION. .......................... 107
FIGURE 36. CORRELATION ANALYSIS BETWEEN DISTANCE AND PERCENTAGE OF CHANGE IN RETINAL THICKNESS.
DISTANCE AT WHICH THE ELECTRODE IS PLACED DOES NOT PREDICT THE CHANGES IN RETINAL THICKNESS. ....... 110
FIGURE 37. CHANGE OF RETINAL THICKNESS DURING STIMULATION OVERTIME. RETINAL THICKNESS CHANGES FOR
DIFFERENT CHARGE DENSITY GROUPS AND DIFFERENT FREQUENCIES ARE PLOTTED IN THIS GRAPH. WE CAN
OBSERVE HOW CHARGE DENSITY AND FREQUENCY ARE NOT THE AFFECTING FACTOR FOR WHEN CHANGES IN
RETINAL THICKNESS START TO BE OBSERVED. ONLY DISTANCE IS THE IMPORTANT FACTOR. .................................... 111
FIGURE 38. RESULTS GROUP 1. STIMULATION PARAMETERS ARE: CHARGE DENSITY 0.92 MCCM2, FREQUENCY 300 HZ,
DURATION OF STIMULATION 30 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE OF THE RABBIT’S EYE DURING ELECTRODE INSERTION. TOP MIDDLE IMAGE. FUNDUS IMAGE SHOWING
xii
ELECTRODE PLACEMENT AFTER IT WAS ADVANCED CLOSE TO THE RETINA. TOP RIGHT IMAGE. FA TAKEN AFTER
STIMULATION WHERE NO VESSEL LEAKAGE IS OBSERVED. BOTTOM LEFT IMAGE. OCT IMAGE TAKEN BEFORE
STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE TAKEN 15 MINUTES INTO
STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT IMAGE TAKEN 15
MINUTES AFTER STIMULATION WAS OVER. NO CHANGES IN RETINAL THICKNESS WERE OBSERVED IN THIS GROUP.
.................................................................................................................................................................................................... 112
FIGURE 39. RESULTS GROUP 2. STIMULATION PARAMETERS ARE: CHARGE DENSITY 1.22 MCCM2, FREQUENCY 100 HZ,
DURATION OF STIMULATION 100 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE OF THE RABBIT’S EYE DURING ELECTRODE INSERTION. BOTTOM LEFT IMAGE. OCT IMAGE TAKEN BEFORE
STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE TAKEN 15 MINUTES INTO
STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT IMAGE TAKEN 15
MINUTES AFTER STIMULATION WAS OVER. NO CHANGES IN RETINAL THICKNESS WERE OBSERVED IN THIS GROUP.
.................................................................................................................................................................................................... 113
FIGURE 40. RESULTS GROUP 3. STIMULATION PARAMETERS ARE: CHARGE DENSITY 1.22 MCCM2, FREQUENCY 200 HZ,
DURATION OF STIMULATION 45 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE OF THE RABBIT’S EYE DURING ELECTRODE INSERTION. TOP MIDDLE IMAGE. FUNDUS IMAGE SHOWING
ELECTRODE PLACEMENT AFTER IT WAS ADVANCED CLOSE TO THE RETINA. TOP RIGHT IMAGE. FA TAKEN AFTER
STIMULATION WHERE NO VESSEL LEAKAGE IS OBSERVED. BOTTOM LEFT IMAGE. OCT IMAGE TAKEN BEFORE
STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE TAKEN 15 MINUTES INTO
STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT IMAGE TAKEN 15
MINUTES AFTER STIMULATION WAS OVER. CHANGES IN RETINAL THICKNESS ARE SHOWN WITH BLUE ARROWS. .. 114
FIGURE 41. RESULTS GROUP 4. STIMULATION PARAMETERS ARE: CHARGE DENSITY 1.22 MCCM2, FREQUENCY 300 HZ,
DURATION OF STIMULATION 30 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE OF THE RABBIT’S EYE DURING ELECTRODE INSERTION. TOP MIDDLE IMAGE. FUNDUS IMAGE SHOWING
ELECTRODE PLACEMENT AFTER IT WAS ADVANCED CLOSE TO THE RETINA. TOP RIGHT IMAGE. FA TAKEN AFTER
STIMULATION WHERE NO VESSEL LEAKAGE IS OBSERVED. BOTTOM LEFT IMAGE. OCT IMAGE TAKEN BEFORE
STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE TAKEN 15 MINUTES INTO
STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT IMAGE TAKEN 15
MINUTES AFTER STIMULATION WAS OVER. CHANGES IN RETINAL THICKNESS ARE SHOWN WITH BLUE ARROWS. .. 115
FIGURE 42. RESULTS GROUP 5. STIMULATION PARAMETERS ARE: CHARGE DENSITY 1.63 MCCM2, FREQUENCY 20 HZ,
DURATION OF STIMULATION 100 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE SHOWING ELECTRODE PLACEMENT AFTER IT WAS ADVANCED CLOSE TO THE RETINA. BOTTOM LEFT IMAGE.
OCT IMAGE TAKEN BEFORE STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE
TAKEN 15 MINUTES INTO STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT
IMAGE TAKEN 15 MINUTES AFTER STIMULATION WAS OVER. NO CHANGES IN RETINAL THICKNESS WERE OBSERVED
IN THIS GROUP. ......................................................................................................................................................................... 115
FIGURE 43. RESULTS GROUP 6. STIMULATION PARAMETERS ARE: CHARGE DENSITY 1.63 MCCM2, FREQUENCY 100 HZ,
DURATION OF STIMULATION 100 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE OF THE RABBIT’S EYE DURING ELECTRODE INSERTION. TOP MIDDLE IMAGE. FUNDUS IMAGE SHOWING
ELECTRODE PLACEMENT AFTER IT WAS ADVANCED CLOSE TO THE RETINA. BOTTOM LEFT IMAGE. OCT IMAGE
TAKEN BEFORE STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE TAKEN 15
MINUTES INTO STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT IMAGE
TAKEN 15 MINUTES AFTER STIMULATION WAS OVER. CHANGES IN RETINAL THICKNESS ARE SHOWN WITH BLUE
ARROWS. ................................................................................................................................................................................... 116
FIGURE 44. RESULTS GROUP 7. STIMULATION PARAMETERS ARE: CHARGE DENSITY 1.63 MCCM2, FREQUENCY 200 HZ,
DURATION OF STIMULATION 45 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE SHOWING ELECTRODE PLACEMENT AFTER IT WAS ADVANCED CLOSE TO THE RETINA. BOTTOM LEFT IMAGE.
OCT IMAGE TAKEN BEFORE STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE
TAKEN 15 MINUTES INTO STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT
IMAGE TAKEN 15 MINUTES AFTER STIMULATION WAS OVER. CHANGES IN RETINAL THICKNESS ARE SHOWN WITH
BLUE ARROWS. ......................................................................................................................................................................... 117
FIGURE 45. RESULTS GROUP 8. STIMULATION PARAMETERS ARE: CHARGE DENSITY 1.63 MCCM2, FREQUENCY 300 HZ,
DURATION OF STIMULATION 30 MINUTES, BIPHASIC, CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE. FUNDUS
IMAGE OF THE RABBIT’S EYE DURING ELECTRODE INSERTION. TOP MIDDLE IMAGE. FUNDUS IMAGE SHOWING
ELECTRODE PLACEMENT AFTER IT WAS ADVANCED CLOSE TO THE RETINA. TOP RIGHT IMAGE. FA TAKEN AFTER
xiii
STIMULATION WHERE NO VESSEL LEAKAGE IS OBSERVED. BOTTOM LEFT IMAGE. OCT IMAGE TAKEN BEFORE
STIMULATION AND ELECTRODE WAS PLACED. BOTTOM MIDDLE IMAGE. OCT IMAGE TAKEN 15 MINUTES INTO
STIMULATION. YELLOW ARROW SHOWS THE ELECTRODE TIP. BOTTOM RIGHT IMAGE. OCT IMAGE TAKEN 15
MINUTES AFTER STIMULATION WAS OVER. CHANGES IN RETINAL THICKNESS ARE SHOWN WITH BLUE ARROWS. .. 117
FIGURE 46. SUMMARY OF RETINAL DAMAGE FOR ALL DIFFERENT GROUPS EVALUATED. RETINAL DAMAGE WAS OBSERVED
AT 1.63 MC.CM-2 (333, 200, 100 HZ) AND 1. 22 MC.CM-2 (333, 200 HZ). RED DOT SHOWS THE PARAMETERS
USED FOR ARGUS II. ................................................................................................................................................................ 119
FIGURE 47. SCATTER PLOT SHOWING THE CORRELATION ANALYSIS BETWEEN FREQUENCY (LEFT IMAGE) AND CHARGE
DENSITY (RIGHT IMAGE) VS. PERCENTAGE OF CHANGE IN RETINAL THICKNESS. IT IS OBSERVED THAT BOTH
VARIABLES ARE PREDICTORS OF THE PERCENTAGE CHANGE OF RETINAL THICKNESS. ................................................. 120
FIGURE 48. BOX PLOT SHOWING THE SUMMARY RESULTS OF THE EIGHT GROUPS TESTED. ON EACH BOX, THE CENTRAL
MARK INDICATES THE MEDIAN, AND THE BOTTOM AND TOP EDGES OF THE BOX INDICATE THE 25TH AND 75TH
PERCENTILES, RESPECTIVELY. THE WHISKERS EXTEND TO THE MOST EXTREME DATA POINTS NOT CONSIDERED
OUTLIERS, AND THE OUTLIERS ARE PLOTTED INDIVIDUALLY USING THE '+' SYMBOL. ................................................. 123
FIGURE 49. OCT IMAGES OF THE RETINA DURING ELECTRICAL STIMULATION AND FOLLOW-UPS. STIMULATION
PARAMETERS OF THIS ANIMAL ARE: CHARGE DENSITY 0.92 MC/CM
2
, FREQUENCY 333 HZ, DURATION OF
STIMULATION 30 MINUTES, BIPHASIC CATHODIC FIRST 1 MS PULSES. TOP LEFT IMAGE WAS TAKEN BEFORE
STIMULATION WAS STARTED. MIDDLE LEFT IMAGE WAS TAKEN 15 MINUTES INTO STIMULATION. BOTTOM LEFT
WAS TAKEN 15 MINUTES AFTER STIMULATION WAS STOPPED. TOP RIGHT IMAGE IS FROM THE FIRST FOLLOW UP 3
DAYS AFTER STIMULATION. MIDDLE RIGHT WAS TAKEN 9 DAYS AFTER STIMULATION. BOTTOM RIGHT WAS TAKEN
TWELVE DAYS AFTER STIMULATION. YELLOW ARROW INDICATES THE TIP OF THE STIMULATING ELECTRODE. BLUE
ARROWS INDICATE THE AREA OF RETINAL THICKNESS INCREMENT AFTER STIMULATION. ........................................ 124
FIGURE 50. PERCENTAGE OF CHANGE OF RETINAL THICKNESS DURING STIMULATION OVERTIME. RETINAL THICKNESS
CHANGES FOR DIFFERENT CHARGE DENSITY GROUPS AND DIFFERENT FREQUENCIES ARE PLOTTED IN THIS GRAPH.
CHANGES IN RETINAL THICKNESS REMAINED THROUGHOUT THE TWELVE DAYS FOR ALL GROUPS. .......................... 126
FIGURE 51. SPATIAL THRESHOLD MAPS FOR THREE PULSE DURATIONS THAT COVER THE GAMUT OF RESPONSE TYPES:
DIRECT RGC STIMULATION (0.06 MS, LEFT COLUMN), COMBINED GANGLION AND BIPOLAR CELL STIMULATION (1
MS, MIDDLE COL- UMN), AND BIPOLAR CELL STIMULATION (25 MS, RIGHT COLUMN) IN RD AND WILD-TYPE (WT)
RATS. IN ALL CASES, 0.06-MS PULSES PROVIDED GOOD SELECTIVITY FOR LOCAL SOMATA OVER PASSING AXONS, 1-MS
PULSES PROVIDED POOR SELECTIVITY AND 25-MS PULSES PRODUCED FOCAL RESPONSES. THE BOTTOM ROW SHOWS
BACKGROUND-SUBTRACTED GCAMP5G RESPONSES TO SUPRATHRESHOLD STIMULI FOR 30 UM-DIAMETER
ELECTRODES (BLACK CIRCLES). IMAGE TAKEN FROM [141]. .......................................................................................... 134
FIGURE 52. THE EFFECTS OF AMPLITUDE AND FREQUENCY ON APPARENT BRIGHTNESS AND SIZE. (A, C) BRIGHTNESS
AND APPARENT SIZE AS A FUNCTION OF NORMALIZED (RELATIVE TO THRESHOLD) AMPLITUDE FOR NINE
ELECTRODES. (B, D) BRIGHTNESS AND APPARENT SIZE AS A FUNCTION FREQUENCY FOR THE SAME NINE
ELECTRODES. EACH ELECTRODE’S DATA ARE FIT WITH THE BEST-FIT LINEAR REGRESSION. FIGURE TAKEN FROM
[80]. .......................................................................................................................................................................................... 136
FIGURE 53. DRAWINGS FROM SINGLE ELECTRODES OVERLAYED ON ELECTRODE ARRAY FOR EACH ONE OF THE PATIENTS
TESTED. RED CIRCLES INDICATES STIMULATING ELECTRODE, BLUE SQUARE MARKS THE APPROXIMATE LOCATION OF
THE FOVEA AND BLACK MARKS REPRESENTS THE SUBJECTS DRAWING TO THAT SPECIFIC ELECTRODE. ................... 137
FIGURE 54. SHAPES PRESENTED TO PATIENTS DURING TACTILE CONTROL EXPERIMENT. .................................................. 142
FIGURE 55. LEFT IMAGE. EXAMPLE OF AN ARGUS II SUBJECT DRAWING TASK. RIGHT IMAGE. SUBJECTS DRAWING
RECORDED BY A TOUCHSCREEN MONITOR. .......................................................................................................................... 143
FIGURE 56. PATTERN STIMULATION GUI WHERE STIMULATION IS CONTROLLED AND PARAMETERS CAN BE VARIED. .. 144
FIGURE 57. LEFT IMAGE. FUNDUS PHOTOGRAPH. RIGHT IMAGE. BIG RED CIRCLE IS MARKING THE OPTIC NERVE, AND
SMALL CIRCLE IS MARKING THE FOVEAL REGION. ............................................................................................................... 146
FIGURE 58. LEFT IMAGE IS A FUNDUS IMAGE OF THE RETINA OF SUBJECT 1 IMPLANTED EYE. OPTIC NERVE AND
ELECTRODE ARRAY ARE USED AS LANDMARKS TO FIND THE ANGLE AND LOCATION OF THE FOVEA. RIGHT IMAGE
SHOWS IN A BIG RED CIRCLE THE OPTIC NERVE AND A SMALL RED CIRCLE THE FOVEA REGION. THE ELECTRODE
ARRAY IS OVERLAPPED WITH THE ARRAY GENERATED BY MATLAB FOR COMPARISON. ................................................ 147
FIGURE 59. DRAWING FROM 12 SINGLE ELECTRODES ARE SHOWN IN THIS IMAGE. IMAGES WITH BLACK BACKGROUND
SHOW THE STIMULUS PRESENTED TO THE PATIENT, THIS IMAGE IS TILTED TO REPRESENT THE TILT OF THE ARRAY
IN THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE PATIENT’S DRAWING WHEN THAT STIMULUS WAS
PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE FOVEA ................................................................... 148
xiv
FIGURE 60. DRAWING FROM 11 DIFFERENT PATTERNS PRESENTED TO SUBJECT 1. IMAGES WITH BLACK BACKGROUND
SHOW THE PATTERN PRESENTED TO THE PATIENT, THIS IMAGE IS TILTED TO REPRESENT THE TILT OF THE ARRAY IN
THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE PATIENT’S DRAWING WHEN THAT STIMULUS WAS
PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE FOVEA ................................................................... 149
FIGURE 61. LEFT IMAGE IS A FUNDUS IMAGE OF THE RETINA OF SUBJECT 2 IMPLANTED EYE. OPTIC NERVE AND
ELECTRODE ARRAY ARE USED AS LANDMARKS TO FIND THE ANGLE AND LOCATION OF THE FOVEA. RIGHT IMAGE
SHOWS IN A BIG RED CIRCLE THE OPTIC NERVE AND A SMALL RED CIRCLE THE FOVEA REGION. THE ELECTRODE
ARRAY IS OVERLAPPED WITH THE ARRAY GENERATED BY MATLAB FOR COMPARISON. ................................................ 150
FIGURE 62. GREEN IMAGES ARE THE SHAPES PRESENTED TO THE SUBJECT. SUBJECT WAS ASKED TO FEEL THEM WITH
BOTH HANDS AND REPORT IN THE TOUCHSCREEN MONITOR THE SHAPE OF THE OBJECT PRESENTED. THE WHITE
IMAGE IS THE PATIENT’S RESPONSE TO THE SHAPE. SHE WAS ABLE TO DRAW ALL SHAPES CONSISTENTLY AND
ACCURATELY. ............................................................................................................................................................................ 151
FIGURE 63. TOP LEFT ARRAY SHOWS 21 X’S REPRESENTING THE 21 ELECTRODES IN WHICH THIS SUBJECT REPORTED
NOT SEEING PHOSPHENES WHEN STIMULATED. 9 EXTRA ELECTRODES WERE RANDOMLY PICKED TO SHOW THEIR
RESULTS. IMAGES WITH BLACK BACKGROUND SHOW THE STIMULUS PRESENTED TO THE PATIENT, THIS IMAGE IS
TILTED TO REPRESENT THE TILT OF THE ARRAY IN THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE
PATIENT’S DRAWING WHEN THAT STIMULUS WAS PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE
FOVEA ........................................................................................................................................................................................ 153
FIGURE 64. DRAWING FROM 12 DIFFERENT PATTERNS PRESENTED TO SUBJECT 2. IMAGES WITH BLACK BACKGROUND
SHOW THE PATTERN PRESENTED TO THE PATIENT, THIS IMAGE IS TILTED TO REPRESENT THE TILT OF THE ARRAY IN
THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE PATIENT’S DRAWING WHEN THAT STIMULUS WAS
PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE FOVEA. THE FIRST TWO ROWS ARE THE RESULTS
OF THE GROUPING EXPERIMENT AND THE LAST TWO ARE THE RESULTS OF PATTERN EXPERIMENTS. ...................... 154
FIGURE 65. LEFT IMAGE IS A FUNDUS IMAGE OF THE RETINA OF SUBJECT 3 IMPLANTED EYE. THIS SUBJECT WAS
IMPLANTED IN THE RIGHT EYE. OPTIC NERVE AND ELECTRODE ARRAY ARE USED AS LANDMARKS TO FIND THE
ANGLE AND LOCATION OF THE FOVEA. RIGHT IMAGE SHOWS IN A BIG RED CIRCLE THE OPTIC NERVE AND A SMALL
RED CIRCLE THE FOVEA REGION. THE ELECTRODE ARRAY IS OVERLAPPED WITH THE ARRAY GENERATED BY MATLAB
FOR COMPARISON. ................................................................................................................................................................... 155
FIGURE 66. GREEN IMAGES ARE THE SHAPES PRESENTED TO THE SUBJECT. SUBJECT WAS ASKED TO FEEL THEM WITH
BOTH HANDS AND REPORT IN THE TOUCHSCREEN MONITOR THE SHAPE OF THE OBJECT PRESENTED. THE WHITE
IMAGE IS THE PATIENT’S RESPONSE TO THE SHAPE. SHE WAS ABLE TO DRAW ALL SHAPES CONSISTENTLY AND
ACCURATELY. ............................................................................................................................................................................ 156
FIGURE 67. TOP LEFT ARRAY SHOWS 6 RED X’S REPRESENTING THE 6 ELECTRODES IN WHICH THIS SUBJECT REPORTED
NOT SEEING PHOSPHENES WHEN STIMULATED AND 9 YELLOW X’S REPRESENTING ELECTRODES BLOCKED BY THE
SYSTEM FOR SAFETY REASONS. 9 EXTRA ELECTRODES WERE RANDOMLY PICKED TO SHOW THEIR RESULTS. IMAGES
WITH BLACK BACKGROUND SHOW THE STIMULUS PRESENTED TO THE PATIENT, THIS IMAGE IS TILTED TO
REPRESENT THE TILT OF THE ARRAY IN THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE PATIENT’S
DRAWING WHEN THAT STIMULUS WAS PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE FOVEA
.................................................................................................................................................................................................... 157
FIGURE 68. DRAWING FROM 12 DIFFERENT PATTERNS PRESENTED TO SUBJECT 3. IMAGES WITH BLACK BACKGROUND
SHOW THE PATTERN PRESENTED TO THE PATIENT, THIS IMAGE IS TILTED TO REPRESENT THE TILT OF THE ARRAY IN
THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE PATIENT’S DRAWING WHEN THAT STIMULUS WAS
PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE FOVEA. .................................................................. 158
FIGURE 69. FUNDUS IMAGE OF SUBJECT 4. PATIENT WAS IMPLANTED IN HIS LEFT EYE. RIGHT IMAGE SHOWS A BIG RED
CIRCLE SHOWING THE OPTIC NERVE AND A SMALL RED CIRCLE SHOWING THE FOVEA REGION. THE ELECTRODE
ARRAY IS OVERLAPPED WITH THE ARRAY GENERATED BY MATLAB FOR COMPARISON. ................................................ 160
FIGURE 70. SHAPE RESPONSES FOR SUBJECT 4. GREEN IMAGES ARE THE SHAPES PRESENTED TO THE SUBJECT. SUBJECT
WAS ASKED TO FEEL THEM WITH BOTH HANDS AND REPORT IN THE TOUCHSCREEN MONITOR THE SHAPE OF THE
OBJECT PRESENTED. THE WHITE IMAGE IS THE PATIENT’S RESPONSE TO THE SHAPE. HE WAS ABLE TO DRAW ALL
SHAPES CONSISTENTLY AND ACCURATELY. .......................................................................................................................... 161
FIGURE 71. TOP LEFT ARRAY SHOWS 30 X’S REPRESENTING THE 30 ELECTRODES IN WHICH THIS SUBJECT REPORTED
NOT SEEING PHOSPHENES WHEN STIMULATED. 9 EXTRA ELECTRODES WERE RANDOMLY PICKED TO SHOW THEIR
RESULTS. IMAGES WITH BLACK BACKGROUND SHOW THE STIMULUS PRESENTED TO THE PATIENT, THIS IMAGE IS
TILTED TO REPRESENT THE TILT OF THE ARRAY IN THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE
xv
PATIENT’S DRAWING WHEN THAT STIMULUS WAS PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE
FOVEA ........................................................................................................................................................................................ 162
FIGURE 72. DRAWING FROM 12 DIFFERENT PATTERNS PRESENTED TO SUBJECT 4. IMAGES WITH BLACK BACKGROUND
SHOW THE PATTERN PRESENTED TO THE PATIENT, THIS IMAGE IS TILTED TO REPRESENT THE TILT OF THE ARRAY IN
THE RETINA. IMAGES IN WHITE BACKGROUND SHOW THE PATIENT’S DRAWING WHEN THAT STIMULUS WAS
PRESENTED. THE RED CIRCLE REPRESENTS THE LOCATION OF THE FOVEA. .................................................................. 163
FIGURE 73. FUNDUS IMAGE OF SUBJECT 5. PATIENT WAS IMPLANTED IN HIS LEFT EYE WITH THE ARGUS II SYSTEM.
RIGHT IMAGE SHOWS A BIG RED CIRCLE SHOWING THE OPTIC NERVE AND A SMALL RED CIRCLE SHOWING THE FOVEA
REGION. THE ELECTRODE ARRAY IS OVERLAPPED WITH THE ARRAY GENERATED BY MATLAB FOR COMPARISON. .. 164
FIGURE 74. SUBJECT 5 RESPONSES TO SINGLE AND PATTERN STIMULATION. IMAGES WITH BLACK BACKGROUND SHOW
THE STIMULUS PRESENTED TO THE PATIENT, THIS IMAGE IS TILTED TO REPRESENT THE TILT OF THE ARRAY IN THE
RETINA. IMAGES IN WHITE BACKGROUND SHOW THE PATIENT’S DRAWING WHEN THAT STIMULUS WAS PRESENTED.
DURING THIS EXPERIMENT, DIFFERENT PATTERNS AND SINGLE ELECTRODE STIMULATION WERE EVALUATED. THE
RED CIRCLE REPRESENTS THE LOCATION OF THE FOVEA .................................................................................................. 166

1
Abstract
Several incurable diseases result in blindness for hundreds of thousands of
individuals each year. Aged related macular degeneration (AMD) and Retinitis
pigmentosa (RP) are two degenerative diseases that affect the photoreceptor cells
causing the retina to lose its ability to translate light into electrical signals. RP is
caused by mutations in over 100 genes that result in the degeneration of rod and
cone receptors and the retinal pigmented epithelium (RPE). Patients will lose their
night vision due to the degeneration of the rods, and their central vision in the later
stages of the disease when macular cones start to degenerate. The overall incidence
of retinitis pigmentosa is 1 in 4000 live births worldwide. AMD causes loss of vision
in the center of the visual field. AMD is the most common cause of irreversible
blindness in elderly individuals worldwide. Studies have shown that 30 percent of
the ganglion cells and 78 percent of the inner nuclear layer cells remain largely
intact in retinas with severe RP and AMD, thus allowing the retina to be electrically
stimulated to restore a sense of vision.

Retinal prostheses are being studied as a method to restore vision. They
have demonstrated the capability to elicit the sensation of light and to give subjects
more independence in their day to day activities. The results are remarkable
because RP patients before implantation had bare or no light perception. Patients
implanted with the Argus II system, in their 5 year follow up are able to perform
2
tasks such as following a white line on the floor and finding a door in a room and
walking towards it.

Some technical issues remain unresolved, clinical studies of epiretinal
implants reveal limitations in the ability of patients to determine the orientation of
grating (used to measure visual acuity), and those who can recognize letters take
more than 40 s to do so. High resolution retinal prostheses will require close
proximity between the electrode array and the retina to maintain low stimulus
thresholds and to minimize the power required by the system, smaller electrodes
to accomplish precise activation of retinal cells to elicit a small visual phosphenes
that can serve as a building block for the pattern, and new stimulation paradigms
that will provide better control over the spatial patterns of activation.

Currently a custom tack is used to attach the electrode array to the retina
making it a challenge to control the electrode-retina distance overtime. During this
report we study and present a novel attachment method based on disintegrin-
integrin binding for retinal tissue. We developed a reliable technique to lase and
produce an active area on the silicone surface, we found a reliable sterilization
method for bioactive materials and we developed a new and reversible method of
attachment to retinal tissue.

Human subject testing has shown patients often see large, elongated
phosphenes due to stimulation of RGC axon bundles. Smaller electrodes will allow
3
more focal stimulation but will also require higher charge densities to have the
same neural activation. It is unclear if these new high charge densities are safe for
the tissue and this is why we developed a new method that allowed us to study the
effects of electrical stimulation on the retina in an in vivo model during the
stimulation period. This is important because it allowed us to study the retinal
response in its natural conditions. In addition, this technique allowed us to study
the long-term effects of retinal stimulation during survival experiments. We tested
the safety of high charge density stimulation in the retina, which is a main factor
that will need to be used in the future during the design of a high-resolution retinal
prosthesis.

4
CHAPTER 1
Introduction
1.1. Eye Overview
The eyes are a very sophisticated organ that helps us process the world. Vision
is a complex chain of events. Light passes through the optical media of the eye
(cornea, aqueous, crystalline lens, and vitreous) and stimulates the photoreceptors
that send the signal to the brain through the visual path. (Figure 1) The brain then
recognizes the image, interprets what has been seen, and responds accordingly [1].
When we spot an object in the environment, the object reflects and scatters the
light waves into our eyes [2]. These light waves pass through the cornea, which is
responsible for two-thirds of the eye’s focusing power. The pupil adjusts
accordingly to the level of light with the help of the iris sphincter and the dilator
muscles. The light continues its path through the crystalline lens, which contracts
or dilates to adjust the focus of the image with the help of the cillary muscle. The
encoded light waves are focused onto the retina as a map of bright and dim colored
dots, known as the retinal image [3].

5

Figure 1. Cross section of the human eye. When light strikes the eye, it passes through
the cornea and the lens to form an image to the retina. Visual information is
transmitted by the optic nerve to the higher visual centers of the brain

1.2 Retina - Anatomy
The retina is a light-sensitive layer of tissue that lines the posterior two –
thirds of the eye, and it is in charge of processing the light energy and relaying it to
the visual cortex for visual perception [4].

The retina is formed of nine different layers; three of them are composed of
very important elements for the transmission of the nerve impulse to the brain
(Figure 2).

6

Figure 2. Retina structure. Schematic of the different cell types and retinal layers
encountered in the retina. The retina converts light into electrical signals that are
carried by the optic nerve to the brain. Process starts at the photoreceptors layer, and
then passed through each layer of the retina to the ganglion cells. Ganglion cells
transmit the action potentials to the cortex for further processing.

1.2.1 Photoreceptor layer
Human eye contains two different types of photoreceptos: rods and cones.
The ratio of rodes to cones is 20:1. They can be distinguished by their morphology
and their behavior in response to light.

Rods are primarily in charge of night vision or vision in dim illumination,
this is why they are highly sensitive to light. They are mostly located in the
periphery of the retina, also making them responsible for peripheral vision. The
rod receptors contain rhodopsin, which its job is to begin the process that produces
the phototransduction cascade.
7
Cones are predominantly located in the macula, which is the area of the
retina in charge of central vision and high-resolution vision. Cones are extremely
tuned to mediate color and spatio-temporal vision. There are three different types
of cones in human retina (red, green and blue), each kind having different but
overlapping spectral sensitivities, that allows us to distinguish small color
differences. Cones synapse with bipolar cells and horizontal cells in the outer
plexiform layer of the retina [5].

1.2.2 Bipolar layer
The bipolar layer relies the information from the photoreceptors to the
ganglion cell layer. The bipolar cells occupy a very important position in the retina,
because all information coming from the photoreceptors and going to the ganglion
cells have to pass through them, which it means they are associated with the direct
and indirect paths of communication. Bipolar cells do not fire impulses, but will
respond to depolarization and hyperpolarization by increasing or decreasing the
amount of transmitter released from the cell’s terminal.

Signal transfer between retinal layers relies on lateral communication
between retinal cells. There are several cell types and neural connections that help
with lateral communication but the major cell types are amacrine and horizontal
cells.
8
Horizontal cells are second order, mainly inhibitory cells. There are three
types of horizontal cells. The input to horizontal cells derives from the rods and
cones response to light, and depending on the type of horizontal cells it can make
contact with many cones over broad retinal areas to ensure optimal retinal
sensitivity over the entire intensity range.

The main purpose of amacrine cells is to provide a mechanism of transfer of
signals within bipolar cells and with ganglion cells. As the horizontal cells, they also
provide a mechanism of lateral communication between retinal cells. They receive
their input from bipolar cells, ganglion cells and other types amacrine cells. There
are 30 different types of amacrine cells; they have been classified depending of
their morphology, physiology properties and pharmacological criteria [6].

1.2.3 Ganglion cell layer
It is the last layer and it is responsible for carrying information to higher-
order visual centers. Ganglion cells take the graded information input and
translate it into a train of actions potentials that travels throughout their axons to
the optic nerve. The ganglion cell axons form the nerve fiber layer, which converges
at the optic nerve head to form the optic disk. There are up to 25 different types of
ganglion cells, classified by their morphological characteristics and their
physiological properties [7].

9
Other important cells that we can find in the retina are:

1.2.4 Muller Cells
Muller cells are the major type of glial cells found in the retina. They are the
architectural support of the retina, stretching radially across the thickness of the
retina and the limits of the retina at the outer and inner limiting membrane
respectively. They control the composition of the extracellular space fluid by
mediating transcellular ion, water and bicarbonate transport. They are involved in
the synaptic activity in the inner retina by uptaking glutamate. They protect the
neurons from mechanical trauma by acting as a compliant embedding for the
neurons [8]. Muller cells respond to every alteration of the retina (retinal trauma,
ischemia, retinal detachment, glaucoma, diabetic retinopathy, etc) by buffering
elevated potassium levels, uptaking excess glutamate that is neurotoxic specially to
the inner retina, releasing antioxidants and producing neurotrophic factors, growth
factors, cytokines and erythropoietin that protect photoreceptors and neurons
from cell death. Gliosis of muller cells have a neuroprotective function, especially
early after injury [9].

1.3 Retina –visual pathway
Retina layers are transparent with the exception of the Retinal pigment
epithelial cells (PPE). When light strikes the eye; it passes through all the layers
until it hits the photoreceptors layer. RPE contains a black pigment call melanin
10
that absorbs the light that passed through the retina, keeping it from being
reflected back [10].

The outer segments of the photoreceptors cells are filled with light-
absorbing visual pigments. Photons are absorbed by the visual pigments, which it
triggers a change in ion fluxes, an enzyme called cGMP phosphodiesterase gets
activated, reducing the amount of cGMP in the cytoplasm. This reduction causes the
cGMP-gated channels to close, hyperpolarizing the photoreceptors.
Hyperpolarization of the cell causes closure of the voltage-gated calcium channels.
The decrease in calcium levels causes a reduction in the glutamate released by the
photoreceptors.

The bipolar cells that synapse with the photoreceptors sense this reduction,
causing a depolarization of ON center bipolar cells and hyperpolarization of cone
off-center bipolar cells [11]. Bipolar cells do not generate action potentials. They
rely the graded information to the ganglion cells, either directly or through
communication with amacrine cells.

Ganglion cells convert the graded electrical signal into a series of action
potentials that travel through the ganglion cell axons, which form the optic nerve,
to the visual cortex for further processing [2].

11
1.4 Retina diseases
One of the reasons for blindness is that the retina loses its ability to
transduce light into electrical signals. We will talk about two conditions that cause
blindness due to retinal degenerative disorders.

1.4.1 Aged- Related Macular Degeneration (AMD)
Aged related macular degeneration affects the light- sensitive photoreceptor
cells in the macular area (the macula specializes in fine acuity vision, all scotopic
vision and 10 percent of our visual field), making it dysfunctional and leaving the
patients with slight visual distortions to complete loss of central vision (Figure 3).
AMD is the leading cause of severe vision loss in adults over the age of 50.
Approximately 1.75 million people in United States have advanced aged related
macular degeneration, and it is estimated that 30 percent of the population over
the age of 75 have some degree of macular degeneration [12].

AMD is classified in two different groups:
Early or Dry AMD is characterized by having abnormal deposits called
drusens below the RPE, and by focal areas of hypopigmentation and
hyperpigmentation in the RPE and choroid. Late AMD has two forms. Neovascular
AMD is recognized by the growth of blood vessels from the choroid into the RPE
and the subretinal space, causing subretinal hemorrhage that leads to faster vision
loss. Geographic atrophy is the end stage of dry AMD and it is characterized by
12
widespread areas of depigmentation and degeneration of the RPE and atrophy of
the capillary bed of the choroid [7].

Figure 3. Representation of patient’s with normal vision (left) and patient’s with
AMD (Right). AMD can vary from slight visual distortions to complete loss of central
vision.

1.4.2 Retinitis Pigmentosa (RP)
Retinitis pigmentosa is caused by a mutation in over 100 genes. Mutated
genes cause the photoreceptors cell to make unbalanced amounts of protein,
leaving the patients with loss of night vision, constricted visual fields, and complete
blindness in its late stages (Figure 4).

Worldwide, Retinitis pigmentosa affects 1:4000 of the population. It is
estimated that more than 100000 people in the USA is affected by retinitis
pigmentosa inherited from one or both parents [13].
13
RP can be classified by its mode of inheritance, age of onset, fundus
appearance, pattern of functional vision loss or by genetic mutation.
Functional loss can be classified in rod-cone dystrophy, cone-rod dystrophy or
cone- dystrophy. The most common is rod-cone dystrophy where the rod receptors
death happens first and its later followed by cone receptors death.
The most common classification is by genetic classification. In Autosomal recessive
forms of RP there is a loss of expression or function when both copies of a given
gene are mutated. Autosomal dominant forms of RP are caused a gain of function
mutation, where the mutated protein becomes toxic or interferes with the function
of the remaining normal forms of that protein [7].

Figure 4. Representation of patient’s with normal vision (left) and patient’s with
retinitis pigmentosa (Right). Patients with retinitis pigmentosa have loss of night
vision, constricted visual fields, and complete blindness in its late stages.

1.5 Treatments
There is no a cure for AMD or RP, some treatments has been developed to
prevent or slow down the vision loss.
14
1.5.1 Photodynamic Therapy
Therapies for AMD have focused mainly in the treatment of neovascular
form of AMD for being the most aggressive of them. Laser photocoagulation and
photodynamic therapy utilizes laser activation of an intraveneously-administered
compound that damages the endothelial cell of neovessels, leading to thrombosis
[7].

1.5.2 Anti- VEGF
Vascular endothelial growth factor (VEGF) plays an important role in
pathological neovascularization. Anti- VEGF drugs have been studied and
developed as a treatment for neovascular AMD. They have been utilized to inhibit
neovascularization and prevent vision loss. These approaches are only applicable
to 5 percent of all AMD patients [14].

1.5.3 Gene Therapy
Replacement of defective genes seems like a promising treatment for
autosomal recessive forms of RP. The main challenge of this treatment is that RP is
caused by a mutation in over 100 genes, making it very complex to design a
multitude of vectors for each one of those genes [15].

15
1.5.4 RPE Transplantation
Replacement of the RPE cells by a suspension of sheet of cultured RPE cells
have been tested as a treatment for RP and AMD. The transplants have been well
tolerated by the patients, but there is not evidence that the cells of the transplanted
tissue develop synaptic connections. In animal models short term studies shows
improvement by modulating photoreceptor death, but long term studies have been
not as expected [15].

1.5.5 Stem-Cell-Based Therapies
Stem cells have been studied as a possible treatment for their capability of
self-renewal and their potential to develop into many specific cell types. One
strategy is to implant the cells to limit the progress of photoreceptor loss, or to
completely replace the death photoreceptors. This is still in progress [7].

1.6 Visual Prostheses
Visual prostheses have been studied as a method to treat outer retinal
degenerative disorders. Diseases that affect the whole retina or the inner layers of
the retina are not suitable for this approach because for retinal prostheses to work
as a treatment, a large amount of retinal neurons most survive to be electrically
stimulated. Studies have shown that AMD and RP cause death of photoreceptors
cells, but bipolar cells and ganglion cells stay largely intact, with a 78-88% of
bipolar cells survival and 30-48% of ganglion cell survival in patients with RP [16],
16
and 90.5% survival of bipolar cells and 69% survival of ganglion cells in patients
with AMD [17].

1.6.1 History
Visual prostheses have been studied as a method to restore vision for
decades, starting in 1929 when Foerster, a German neurosurgeon, exposed the
occipital pole of one cerebral hemisphere and electrically stimulated the visual
cortex in a sighted patient, reporting that his patient saw a small spot of light [18].
In 1931, Krause & Schum, were able to create a sensation of light by replicating
Foerster’s experiment in a patient who had been blind for over eight year [19].
Similar experiments kept being performed over the decades [20] [21] .

In 1968, Brindley and Lewin implanted the first visual prosthesis, in a blind,
52 year old woman. The extracranial part of the implant consisted of an array of 80
radio receivers encapsulated in silicone rubber, that connected through a cable to
the intracranial part of the implant, which consisted of an array of 80 platinum
electrodes placed in a silicone rubber cap that was placed on her visual cortex
(Figure 5) [22]. Patient reported seeing spots of white light in 39 different locations,
based on the position of the electrode that was being stimulated. Six year after
implantation a new publication reports that the device was still partially functional
[23].

17

Figure 5. Left image. Brindley’s cortical prosthesis shown prior to implantation. Right
image. X-ray of the cortical prosthesis post implantation. The arrow shows the
receiver array, and the arrowhead shows the electrode array.

The amazing work reported by Brindley and Lewin inspired other scientists
to explore this field. In 1970, William Dobelle implanted a 64 platinum disk
electrode array in the right occipital lobe of two completely blind patients. The
computer controlled stimulator presented different pattern stimulus protocols,
simple patterns and letters to both patients. Patients were able to draw the
patterns consistently [24]. Dobelle’s publication reports patients seeing single
phosphenes when multiple electrodes were stimulated, which it made them
understand a more complex explanation than differential brightness was needed
for the issues they were experiencing.

In 1990, electrodes were used to penetrate the cortex throwing some
interesting findings. It was reported that electrodes needed to be 700 microns
away from each other to produce independent phosphenes, and these phosphenes
18
could be produced with electrical current thresholds 10 to 100 times lower than
those produced by stimulation with non penetrating cortical surface electrodes
[25].

The prototype of a visual prosthesis placed in the visual cortex kept being
studied, Although stimulating the visual cortex showed promise, its organization is
very complex, and this is why different locations of the visual pathway, specially
the optic nerve and the retina, started to be explored for the placement of
prostheses.

Different groups have adopted different approaches to investigate these
visual implants but we will focus on retinal implants.

1.7 Retinal Prostheses
A retinal prosthesis consists of multiple components. In most cases, an
external acquisition system is used to capture images from the outside; wearable
computers convert this image into a stimulation pattern, a telemetry system
delivers power and data to the implanted stimulator, the stimulator generates
electrical pulses, and an electrode array delivers the stimulus pulses to the retina.
[26] Placement and complexity of implanted components are the main differences
that distinguish the various systems. In particular, the placement of the
microelectrode array is a critical, defining feature of retinal prostheses. There are
19
three different placements for the retinal prostheses electrode array. Epiretinal
prostheses are attached to the inner surface of the retina, nearest the ganglion cells.
Subretinal prostheses are placed between the retina and the choroid, in the space
where the photoreceptors were prior to degeneration and suprachoroidal
prostheses are implanted between the sclera and choroid, behind the retina [27].
(Figure 6)

Figure 6. Illustration showing the eye structure, the stratified structure of the retina,
and the placement for epiretinal, subretinal, and suprachoroidal prostheses.

1.7.1 Epiretinal prostheses
Epiretinal prostheses refer to implant where the microelectrode array is on
the ganglion cell side. In practice, the electrode array is fixed to the retina with a
tack [28]. This type of placement allows direct stimulation of ganglion cells (the
final output of the retina), which may be advantageous in retinal degeneration
diseases where the retinal network is altered. Another advantage of epiretinal
20
implantation involves the use of the vitreous cavity for implantation and heat
dissipation.[29], [30]. Surgical access is easier, with reduced risk of retinal
detachment. However, fixation of the device to the retinal surface remains a
challenge. [31] Since the array is tacked at only a single location, parts of the array
away from the tack can separate from the retina. The three main groups that have
reported clinical studies of epiretinal prostheses are Second Sight Medical Products
(SSMP), Intelligent Medical Implants (now Pixium), and EpiRet.

1.7.1.1 Second Sight Medical Products
Two devices have been developed by SSMP: Argus I and Argus II. The
external system is virtually the same for both, but the implant is different. The
external system has a small camera placed on a pair of glasses, that captures the
image, and feeds it to an external video processing unit (VPU) for image processing.
The VPU software determines the average brightness of each region in a camera
field of view and uses this information to program stimulation parameters for the
implant. These parameters are encoded into a serial data stream and transmitted
via a radio frequency (RF) signal to the implant. The RF signal also provides
adequate energy to allow the implant to recovery power for operation. [32].
(Figure 7)

Argus I is a first generation epiretinal prosthesis approved for an
investigational trial by the United States Food and Drug Administration (FDA). The
21
main goal of this trial was to demonstrate the safety of long-term retinal
stimulation. The electronics were implanted behind the ear; since the electronics
were based on cochlear implant technology, with a cable running along the temple
to the orbit. A 4x4 electrode array, at the end of the cable, was implanted into the
eye and attached to the retina with a tack. Platinum electrodes were 520 or 260 !m
in diameter, the center to center electrode spacing was 800 !m [33]. The electrodes
were supported on a silicone rubber substrate. Six subjects were implanted
between 2002 and 2004. After some training, patients could use the Argus I to
accomplish simple visual tasks, like localizing a white square on a black computer
display, report the direction of motion of a moving bar, find a black door on a white
wall, and follow a white line on the floor. In general, patients performed these tasks
better with the system on than with the system off [34]. A recent report showed
maintained functionality of the Argus I 10 years after implantation [35]. Overall,
Argus I demonstrated adequate safety and some improvements in visual function,
thus motivating the development of Argus II.

22

Figure 7. Argus epiretinal prostheses: (A) External parts of the Argus system. (B)
Extraocular and intraocular part of the Argus II system. (C) Fundus image of a
patient implanted with the Argus I system. Argus I implant contains 16 electrodes. (D)
Fundus image of a patient implanted with the Argus II system. Argus II implant
contains 60 electrodes.

Argus II is a commercially available device that obtained the CE mark in
2011 and FDA approval as a humanitarian device in 2013. Over 100 devices have
been implanted to date [36]. The electrode array consists of a 6x10 grid of
roughened platinum disks. Each disk is 200 !m diameter and the center-to-center
spacing is 575 !m. Similar to the Argus I, the electrode array is attached to the
retina with a tack [37]. During clinical trials 30 patients were evaluated for a
23
minimum of six months and up to 2.7 year. Multiple tasks were given to the
subjects and their performance was evaluated by comparing the results with the
system ON vs the system OFF. Patients performed statistically better with the
system ON than with the system OFF in object localization (96% of subjects),
motion discrimination (57%), and a discrimination of oriented grating (23%). The
best visual acuity recorded was 20/1260 [38]. In more recent study, thirty patients
in ten different centers in the United States and Europe were evaluated. 24 out of
30 patients remained implanted with functioning Argus II system. Patients were
asked to perform three real-world functional tasks after 5 years of implantation:
Sock sorting, sidewalk tracking and walking direction discrimination tasks. All
patients performed better with the system ON than with the system OFF. This
result supports the long term safety profile and benefit for blind patients [39].

Overall, the Argus II has shown that retinal implants can partially restore
vision and improve people’s quality of life by helping them be more independent.
The main limitations found in Argus II are limited resolution due the relatively
large electrodes, a head-fixed camera that requires patients to scan to change an
image and may result in confusing input by ignoring eye movements, and the use of
the retinal tack, which by design damages the retina and may cause the array to tilt.
[40]

24
1.7.1.2 Epi-Ret

EpiRet3 is a retinal prosthesis designed by the Epi-Ret consortium in
Germany. It is distinguished by the fact that the implant fits entirely inside the eye.
The intraocular component consists of a receiver coil, a receiver chip and
stimulator chip, positioned in the location of the lens, and a electrode array with 25
3D- stimulating electrodes, 25 !m height and 100 !m diameter with a center to
center spacing of 500 !m [41]. The electrodes are formed by electroplating of gold
to achieve a height of 25 !m but are covered with a thin layer of iridium oxide,
which is a better material for neural stimulation.

Six legally blind patients were implanted for a 4-week period in 2006. No
external system was available for home use, but the implant could be activated in
the clinic. During the 4-week period patients were evaluated at day 7, 14 and 27. All
patients reported the occurrence of visual perceptions during stimulation, by
pressing a button or verbally describing the perceptions like dots, line or circles
depending of the stimulation applied. Due to their 3D stimulating electrodes, close
proximity to the retina was achieved and low thresholds
(!"#$""% 73.2 !" !"
!
!"# 7.8 !" !"
!
) were measured during the experiments
[42].

The Epi-Ret device was removed after 4 weeks, but the retinal tack was left
in place to avoid trauma from removal. The electrode array and tack were designed
25
to allow explantation of the array without removal of the tack. Two years after
explantation, patients were re-examined for long-term side effects. No structural
alterations were found, quality of life was consistent with their initial baseline, and
moderate epiretinal gliosis was reported in the area were the tack was implanted
[43].

1.7.1.3 Intelligent Medical Implants
Intelligent Medical Implants (IMI) was founded in 2002 to commercialize
retinal prosthesis technology initially developed at the Fraunhofer Institute. Their
retinal prosthesis design consists of a retinal stimulator with an extraocular coil
attached to the outer wall of the eyeball, and an intraocular multi-electrode array
attached to the retina with a tack, a visual interface (camera, data and energy
transmitter) mounted in a pair of eyeglasses and a pocket processor for image
processing and power supply (Figure 8)[44]. Two wireless links are used: one RF
transmission for power and one infrared link for data. The intraocular component
has an infrared receiver that translates the optical signals into electrical impulses
and sends them to a polyimide 49 contact electrode array with 360 !m diameter
electrodes and an integrated microcable with conducting lines [45]. The pocket
processor included a retinal encoder designed to replicate retinal signal processing.

Twenty subjects were tested between 2003 and 2004 with short-term
electrode array implants (less than 3 hours in a surgery room setting). Nineteen
26
out of the twenty subjects reported light perception during stimulation. Even if
only one electrode was stimulated, subjects described light perceptions as points,
circles, triangles, rectangles, etc. Different colors as white, yellow and blue were
also reported [46].

Based on these results, the implant was developed and four patients were
implanted chronically to test the feasibility and safety of the surgery. Follow ups
were done during a 9 month period, reporting tolerance to the implant. During
stimulation patients were able to distinguish between different points and simple
patterns such as horizontal bars [47].

Intelligent Medical Implants is currently called Pixium and their system IRIS
is in clinical trials. Iris consists of 150 electrodes. Their design allows explantation
of the device with minimal retinal damage. The first implantation was performed
in January 2016, and they expect to enroll a total of 10 patients in their clinical
trials. [48].
27

Figure 8. Epiretinal prosthesis. (A) Schematic of the Intelligent Medical Implant (IMI)
system. (B-C) IMI prototype.

1.7.2 Subretinal Prostheses
Subretinal prostheses refer to the retinal implant where the microelectrode
array is placed behind the retina, in the space previously occupied by the
photoreceptors (Figure 6). Due to degeneration of the photoreceptor layer caused
by RP and AMD [13] [12], this configuration allows the implant to directly
stimulate the bipolar cells, thus taking advantage of the neural processing within
the inner retina. Even though the surgical procedure is more complex than the one
for an epi-retinal prosthesis, it has the advantage of not having to tack the electrode
to the retina but instead keeping the electrode in place by pressure on the retina.
28
The electrode cable is also sutured outside the eye. For the electrode array
placement, a controlled retinal detachment (bleb) is performed in the subretinal
space, increasing the complexity of the surgery, the risk of surgical complications
and limiting the electrode array space to the bleb size. If an active circuit is placed
in the subretinal space, then heat generation becomes an issue, since the retina is
so close to a potential heat source.

Subretinal prostheses can be classified in two different groups based on the
design of the subretinal component: micro-photodiode arrays (MPDAs) or micro-
electrode arrays (MEAs). MEAs are the approach also used by epi-retinal
prostheses already described above. MPDA include light sensitive elements
(photodiodes) on the subretinal component and thus detects the light incident on
the retina. Each photodiode independently converts the luminance level into
electrical stimulus that is delivered to the inner retina [49].The four main groups
that are pursing the subretinal approach are: Optobionics (MPDAs), Retinal
Implant AG (MPDAs), Boston Retinal Implant project (MEAs), and Stanford
optoelectronic retina (MPDA).

1.7.2.1 Optobionics
The ASR (Artificial Silicon Microchip) was implanted in 6 subjects starting in
2001 as part of a phase I clinical trial. The implant design consisted of a 2mm
diameter, 25 !m thick ASR chip. 5000 microphotodiodes were built on top of an
29
electrical ground covering the entire surface. Each pixel was 20x20 !m square with
a 9x9 !m iridium oxide electrode. After subretinal implantation, the implant didn’t
require power supply or data transmission, making it a completely autonomous
device [50]. During an 18-month period, there were no surgical complications.

Patients reported an improvement in their visual capacities (improved
visual acuity, improved contrast and color perception, and enlarged visual fields)
that were not necessarily related to the implant and possibly duty to a
neurotrophic effect caused by the ASR implantation that improved retinal health
[51]. These 6 subjects were followed up to 8 years after implantation [52].
Theoretical analysis of the possible output current of the microphotodiodes
estimates maximum current to be in the picoAmpere range, well below the
microAmperes of current needed for neural activation. This miniscule amount of
current is due to the small amount of light that reaches the back of the eye [53]. As
a result of the ASR trial, the purpose of Optobionics changed from developing a
retinal prosthesis to a therapeutic device capable of rescuing retinal function and
restoring vision of the type that was lost in retinal dystrophies.

42 patients were implanted in a phase II multicenter trial. As a therapeutic
device, implantation was performed in a paramacular location to avoid insertional
injury to the macula and allow neurotrophic rescue of its function. Interim results
showed persistent neurotrophic restoration of visual function [54].

30
1.7.2.2 Retina Implant AG
Retina Implant AG developed a subretinal implant called Alpha IMS which
started clinical trials in 2005 and received the European CE marking in 2013. Two
different implants have been developed. The initial prototype included a
percutaneous connector for provision of power and configuration data, as well as
direct access to a small array of stimulating electrodes. This first prototype was
strictly experimental. The next generation of the device was designed as a
commercial implant, with a wireless power delivery system [55]. The rest of the
components are the same including a subretinal chip 3x3 mm, 17 !m thick, placed
on top of a polyimide foil 53 !m thick that is connected to a subdermal cable. The
subdermal cable connects to either the percutaneous connector (first generation)
or the implanted power module (second generation). Both the connector and
power module are behind the ear, requiring tunneling from the lateral wall of the
orbit to behind the ear. In the 2
nd
generation device, the external system consists of
a transmitter coil, and power pack that allows adjustments to the brightness and
contrast of the perception [56] (Figure 9). The MPDA chip includes 1500 pixels.
Each pixel has a photodiode-amplifier-electrode (titanium nitride) that absorbs the
incoming light, transforms and amplifies it into electrical current for stimulation
via the electrode. The photodiodes are 15 x 30 !m in size with a space between
them of 70 !m. The first device has an additional 16 -50x50 !m or 100x100 !m-
electrodes for direct stimulation (DS). The DS electrodes were connected to the
percutaneous connector via separate traces on the polymide foil. The main purpose
31
of these extra electrodes was to do more elaborate studies of the electrode-retina
interface without the constraints of the MPDA [57].

The first device was implanted in 11 patients. During experiments using DS
electrodes, 6 out of 11 patients reported light perception [58]; when using both DS
electrodes and photodiode-amplifier-electrodes, 2 more subjects reported light
perception. From these 8 subjects that reported light perception, six were able to
detect patterns like lines and letters formed with the direct stimulation electrodes
[59]. 3 out of the 11 subjects were tested only with the MPDA vision and all were
able to detect light. One of the three was able to recognize letters. These three
subjects were implanted for 126 days.

The second device has been implanted in 29 patients. 79% of the patients
reported improvement on daily living activities: mobility and recognition tasks, and
86% of the patients reported improvement in light perception and/or object
recognition. Patients reported seeing the shape of another person’s head, glasses,
house outlines, pavement lines, landmarks, and car lights moving at night[60]. In
some cases, a decrease in functionality was noted at the 3-month follow up due to
technical difficulties, mainly caused for cable breakage due to mechanical stress for
eye movement. Another main issue reported was failure of the hermetic seal
causing corrosion to the chip, which with time led to function loss [61]. The group
reports technical fixes of both issues for future devices: Adding a loop to minimize
32
cable stress and a modified encapsulation technology can lead to longer life
devices.

Figure 9. Subretinal prostheses. (A) Schematic of the Boston retinal implant showing
the secondary coil surrounding the cornea. (B-C) Retinal implant from Retinal
Implant AG called Alpha IMS. It consists of the vision chip (multiphotodiodes array)
on a polyimide foil (both placed subretinally), a power supply cable connecting the
microchip with the receiver coil in a ceramic housing, and the reference electrode
placed subdermally at the temple and retroauricular region.

1.7.2.3 Boston Retinal Implant Project
The MIT-Harvard group that forms the Boston Retinal Implant Project has
had three different retinal prosthesis prototypes.

33
Their first prototype consisted of an epiretinal array [62], and their second
prototype was a subretinal implant that consisted of an electrode array with 15
sputtered iridium oxide contacts, 400 !m in diameter [63]. This device was
implanted in 2008 in 2 yucatan minipigs, but the implant needed to be removed
three and five and a half months after implantation because the conjunctiva over
the device wore through exposing the device [64].

Their third prototype consists of the same components as the second
prototype. It consisted of an external and an internal component. The external
component included a computer controller to select commands for the strength
and duration of the electrical stimulation; commands and power were transmitted
wirelessly to the implanted components of the prosthesis by near-field inductive
coupling. A chip attached outside of the eye decoded the incoming data and
delivered current pulses to an electrode array with 256 sputtered iridium oxide
contacts, which it was implanted into the subretinal space and sutured to the sclera
where it entered the eye (Figure 9). Electrodes were 400 !m diameter. The
increased number of electrodes required an upgraded telemetry system. Their
class E circuit created a carrier field at 6.78 MHz, transmitting up to 30 mW of
power, encodes FSK data at 565 Kbps, using 8b/10b digital encoding and received
LSK encoded data from the chip at 47 Kbps. The hermetic package of the new
device is also curved to be more conformal to the eye, to avoid conjunctival
breakdown problems reported with their second prototype. The hermetic
enclosure design was tested for helium leakage, and measured for hermeticity. A
34
helium leakage rate of 10
!!
!"# 10
!!
!"#$%#&% !! !"# was measured across the
devices giving them an expected life of 5 to 10 years. For clinical trials a camera and
portable image processor will be added to the system [65]. No results have been
reported on the validation of this device.

1.7.2.4 Stanford University
Stanford University developed a MPDA retinal prosthetic system that uses
near infrared (NIR) light projected into the eye. [66]. Their system consists of a
video camera that captures an image from the environment. A pocket computer
processes this image. A near-to-eye projection system (goggles) is used because
ambient light is too dim to produce sufficient photocurrent from the MPDA to
stimulate neurons. This system projects the image into the eye and onto a MPDA
using pulsed NIR (880-915 nm) light. A MPDA is implanted subretinally, one pixel
or module (70 and 140 microns) corresponds to a Iridium oxide electrode (20 and
40 microns in diameter), surrounded by 2-3 photodiodes in series arranged in an
hexagonal pattern. Each pixel of the array converts NIR light into cathodal-
first/anodal-first (2 different prototypes), charge-balance pulses. Each module can
be implanted independently, to avoid large sclerotomy and still be able to cover a
large visual field [67].

This system has not been implanted in humans, so safety and efficacy is yet
to be determined. Single modules were implanted in RCS rats and wild type (WT)
35
rats showing good toleration of the implant in subretinal space, and efficient retinal
stimulation at safe irradiance levels [68]. OCT images show thinning of the retina at
the implant site in wild type rats, but no changes in RCS rats. Their study shows
that when using 140-micron pixel with three diodes, the threshold for anodic
pulses was 4 times lower than cathodic pulses (0.25 ±0.09 !"/!!
!
vs 1.0
±0.3 !"/!!
!
) and 2.7 times lower when using 70 microns pixels with three
diodes. All these values are below ocular safety for NIR radiation [69].

A most recent study performed in rabbits, show the implantation of multiple
modules (up to seven) in the subretinal space. During implantation, a 1.5
retinotomy was performed, and modules were implanted one by one into the
subretinal space. They report that using small rigid modules fits better the
spherical shape of the eyeball compared to a single, wider array. With seven
modules they covered 3.5 mm of the retina, corresponding to 12 degrees of visual
field [70]. Long-term studies are yet to be published.

1.7.3 Suprachoroidal Prostheses
Suprachoroidal prostheses (STS) are implanted in a pocket formed between
the choroid and the sclera (Figure 6). The surgical approach does not involve direct
manipulation of the retina, so potential for surgical trauma to retina is decreased,
compared to epiretinal and subretinal placement [33]. The fact that the electrode is
placed so far from the retina lessens the risk of retinal detachment. Also, because
36
the electrode is placed by the choroid, the increased blood flow of this layer [71]
may help with heat dissipation via convection. On the other hand, suprachoroidal
placement requires higher current intensities for neural stimulation [72] because
the distance between the electrodes and the retinal neurons is greater.

We will review two different STS projects in this chapter: Bionic Vision
Australia Research and, Department of Visual Science- Osaka University.

1.7.3.1 Bionic Vision Australia (BVA)
BVA suprachoroidal implant was implanted in 3 patients from 2012-2014.
Their study included 4 follow-ups for a two-year period [73]. Their implant
consisted of 33 platinum electrodes, where 30 were 600 !m in diameter and 3
were 400 !m in diameter as well as 2 return electrodes, each 2000 !m in diameter
and on the STS array, and a third large return electrode that was implanted behind
the ear. The 13 electrodes that corresponded to the outer ring of the 33 platinum
electrodes were shorted together to study hexagonal stimulation [74], leaving them
with a total of 20 stimulating electrodes and 3 return electrodes (Figure 10). The
electrode array was connected to a percutaneous connector that was anchored to
the skull with titanium screws behind the subject’s ear.

37

Figure 10. Suprachoroidal prostheses. (A) Suprachoroidal implant. This implant
contains 49 electrodes. (B) Bionic Vision Australia (BVA) implant consists of one
remote return, two other return electrodes on the suprachoroidal array, and a chip of
33 platinum stimulating electrodes.

After surgical implantation, all patients developed a subretinal and
suprachoroidal hemorrhage, which resolved in a 55 to 101 days period. In all three
patients, the implant remained functional for over the twelve-month period [75].
All patients reported phosphene percepts during direct stimulation experiments.
All subjects showed improved light localization with the device on vs the device off
and 1 patient achieved an estimated Landolt-C visual acuity of 2.6logMAR
(20/8397) [76].

38
1.7.3.2 Department of Visual Science – Osaka University
A suprachoroidal implant was developed and implanted in two patients in
2011 and tested during a 4-week period [31]. The implant consisted of an electrode
array with 49 platinum electrodes, but only 9 were active since only 9 stimulation
channels were available from the electronics module. The electrodes were 500 !m
in diameter with a center-to-center spacing of 700 !m. The return electrode was a
6 mm long, 500 !m diameter wire and it was placed in the vitreous cavity. The
electronics module was placed behind the ear [53] (Figure 10). Although electronics
were limited to 9 channels, the investigators plan to expand this to 49 channels,
which is why the electrode array had more contacts than could be used.

Phosphenes were reported on 5 of 9 electrodes in one patient and 6 of 9 in
the other. Patients were asked to perform simple visual tasks, including object
discrimination (Both patients), and grasping objects (Only patient 2) [77]. Patients
performed better with the system ON vs the system OFF in all tests. After the 4-
week period, the implant was removed.

A more recent study reports the implantation of a 49 channel STS retinal
prosthesis in three patients and followed up for one year for safety and
effectiveness. The square localization test (P<0.05) and table tests (P<0.05) were
significantly better with the system ON than with the system OFF in patient 3. The
deviation of the walking test was smaller with the system ON than the system OFF
in patient 2 (P<0.01) and patient 3 (P<0.01). Retinal prosthesis was still functional
39
during the one year follow up for all three patients with 36, 34 and 18 electrodes
respectively eliciting phosphenes with currents less than 1 mA. Greater
improvements of visual tasks were observed in patients where the array was
implanted closer to the fovea centralis[78].

In conclusion, Retinal prostheses have steadily evolved from the laboratory
into a medical device with regulatory approval. Different approaches have been
reported as a possibility to restore vision in blind patients. Several of these groups
have been successful at restoring partial vision in patients, allowing them to be
more independent in their day-to-day life. For instance, Second Sight Medical
Products and Retina Implant AG have commercially available devices. These
devices have enabled the acquisition of clinical data that could be used as a
foundation for the development and improvement of future retinal prostheses.

Some technical issues remain unresolved, giving each one of these systems
room for improvement. We will focus in the difficulties that the Argus II system will
need to overcome for the development of a high – resolution retinal implant.

1.8 Problems with retinal prosthesis
If the future will bring the development of a high-resolution retinal
prosthesis, some of the technical issues that will need to be resolved are:

40
1.8.1 Electrode-Retina Distance
Maintaining the electrode array close to the retina is very important for a
successful retinal prosthesis. Chronic implant studies have shown a strong
correlation between electrode-retina distance and electrical threshold; perceptual
thresholds below 1mC.cm-2 were present when the electrodes were in contact with
the macula [40]. Close proximity may become even more critical for a future high-
resolution implant that will use smaller electrodes [79]. Currently a tack is used to
attach the Argus II retinal prosthesis to the retina [36], which it makes it more
difficult to control the electrode array-retina distance and the placement. A
limitation of this fixation method is the inconsistency of positioning the electrode at
the same distance epiretinally.

1.8.2 Electrode size
Fitting 600-1000 electrodes in a 5mm diameter of the retina [66] means
that small electrodes will need to be used (tens of microns). Small electrodes will
focus the stimulus. As an example we can see in the groups reported above that
electrode size makes a difference in visual acuity. Argus II used 200-micron
electrodes and reports a visual acuity of 20/1260, Alpha IMS with smaller
electrodes reports a visual acuity of 20/200.

41
1.8.3 Axonal stimulation
Single electrode stimulation experiments performed in patients implanted
with the Argus II, reported the visualization of elongated percepts when one single
electrode was being stimulated [80] [81]. Ideally, in a retinal prosthesis one
electrode should activate only nearby cells, and should produce small round
percepts[82] . This is important for shape perception. Visual simulations have
shown that to be able to design a high-resolution retinal prosthesis, a total of 600-
1000 electrodes, with dimensions ranging between 50-100 microns in diameter
will be required in a 5mm diameter on the retina [50] [31]. If focal, small, round
percepts are not achieved, the stimulation from multiple electrodes will overlap
due to the proximity of the electrodes and high-resolution vision will not be
accomplished.

1.8.4 Stimulation selectivity
The retina is a complex neural network feature many, parallel information
channels. To truly replicate natural vision would require the ability to activate
selectively these channels. For example, two classes of retinal ganglion cells are
“on” and “off” cells, so named because the cells respond to the onset or offset of
light. Currently, these cells are activated at the same time by electrical stimulation,
an occurrence that would not happen naturally. While implant patients do see light,
it remains an artificial light and not natural in appearance. To selectively activate
different channels, more sophisticated stimulation schemes may be needed.
42
This work seeks to test different experimental techniques that will enable
the improvement of the current technology and could potentially be used in the
development of high-resolution artificial vision. In particular we will focus in
evaluating new experimental models that will help in finding solutions for the first
three issues mentioned above.

43
CHAPTER 2

Disintegrin-Integrin novel attachment to help with
Electrode-Retina Distance

2.1 Background
2.1.1 Importance of retinal distance
Low stimulus thresholds are desired to minimize the power required by the
system, to create local percepts due to the localization of the electric field and
because they have been proven to be the safest and most efficient for retinal
devices. Currently, a major concern is that the current amplitude necessary to elicit
percepts varies throughout time due to variation in electrode-retina distance
because of the instability of position of the electrode array on the retina surface,
neurophysiological changes, or electrochemical changes on the electrode surface
[83].

Visual simulations have shown that to be able to design a high-resolution
retinal prosthesis, the implant will have smaller electrodes to be able to stimulate
44
one ganglion cell with each electrode. A total of 600-1000 electrodes, with
dimensions ranging between 50-100 microns in diameter will be required in a
5mm diameter on the retina [84]. Using smaller electrodes will require higher
charge densities to obtain neural activation, thus low threshold stimulus will
become even more critical for a high-resolution implant.

Chronic implant studies have demonstrated a strong correlation between electrode
– retina distance and electrical threshold (
Figure 11); it was found that 90.3% of electrodes placed in direct contact
with the retina and within 3 mm of fovea centralis had thresholds below 1mC.cm-2
and 80.9% had thresholds below 0.35 mC.cm-2 [40].

Figure 11. The effect of electrode-retina distance. Study performed in patients
implanted with the Argus II system. The mean electrode threshold across all subjects
45
was 206 ±6.3 !" (n=703 electrodes) and the mean electrode-retina distance was
179.6 ±6.5 !" (n=1013 electrodes). [40]

2.1.2 Epiretinal prosthesis – Tack
Using an epiretinal implant means placing the microelectrode array on the
ganglion cell side, which are the final output of the retina, and remains largely
intact in retinal degenerated diseases where the retinal network is altered [16].
However, fixation of the device to the retinal surface remains a challenge.

Tacks were initially developed to treat complicated retinal detachments
[85] [86] and have been used successfully for the fixation of retinal prostheses.
Currently a custom tack is used to attach the electrode array to the retina (Figure
12). A tack hole is located at the heel of the array, near the cable insertion point.
The array is positioned over the macular region and the retinal tack is inserted
piercing the retina, choroid and sclera to anchor the array in place. The tip of the
tack anchors itself outside of the sclera, securing the tack in place [87].

As a method of attachment, tacks have worked very well, but it causes a lot
of injury to the tissue in the tack site provoking a vascular reaction [88], and since
the array is tacked at only a single location, parts of the array away from the tack
can separate from the retina, making it a major challenge to control electrode-
retina distance over time.
46

Figure 12. Retinal Tacking. Left image. Illustration of an Argus I array placed
epiretinally, showing the instrument and technique use to tack the array to the retina.
Right image. Fundus image of a patient implanted with the Argus II array where the
tack is easily visible on the top of the array.

In this chapter we will describe a novel attachment method for retina tissue
that utilizes silicone modified with bioactive molecules.

2.1.3 Integrins
Integrins are a family of cell adhesion receptos that play important roles
during developmental and pathological processes and have only being
characterized at the molecular level for aproximately 30 years [89]. Their overall
purpose is to maintain tissue architecture in a differentiated state, coordinate and
mediate cell movement, adhesion and migration via firm, reversible, and transient
contacts with the extracellular matrix [90]. Integrins must go to rapidly
47
conformational changes to allow the cell to attach and detach from the extracellular
matrix [91].

Integrins are heterodimeric glycoproteins composed of two different
subunits: alpha – α chains and beta – β chains [92]. So far eighteen α- and eight β-
subunits have been described for integrins in human cells; where combined form
the twenty four heterodimers that are part of the integrin family [93]. An
extracellular domain involved in ligand binding, a single transmembrane domain,
and a cytoplasmic domain, which regulates integrin function, forms each subunit.
Integrins bind to a large number of the ECM components, but the most utilized
binding site is the arginine-glycine-aspartic acid (RGD) sequence that is present in
extracelullar molecules like fibronectin (FN) and vitronectin (VN) [94]. (Figure 13)

Integrins exist in three conformational states: an inactive or low affinity
state, a primed or activated high affinity state, and a ligand bound or occupied state
[95]. Even though it is posiible for some integrins to bind during the low affinity
state, to take full advantage of the binding capabilities of the integrins they need to
be in the active high affinity state [95]. Some activated integrins are not displayed
by quiescent tissue, such as av and a5 members, but play an important role in
processes including attachment, invasion and angiogenesis. Angiogenesis is the
process where some biochemical and molecular behaviors form new vessels from
pre-existing vessels, which it plays an important role in wound repair,
inflammation and tumor growth. [96] [97].
48

In humans, integrin subunits (a1, a2, a3 and β1) are present in the Muller
cell [98] foot end processes, which forms the inner limiting membrane of the retina.

Figure 13. Illustration of integrin-disintegrin interaction. Integrins are a family of
transmembrane receptor proteins, that bind to components of the extracellular
matrix. Disintegrins contain a secondary structure known as “RGD adhesive loop”,
which plays a very important role in the interaction with specific integrin receptors.

49
2.1.4 Disintegrins
Disintegrins are a specific family of proteins that bind and inhibit integrin
function [99]. They are synthesized in the venom glands of snakes from the
hemotoxic Viperidae family [100]. Although disintegrins are found in toxic venom,
they are not responsible for the hamful effects of venom, and their main purpose is
to inhibit integrin function while other toxins do their job.

Disintegrins are small, rich in disulfide bonds, and contains a secondary
structure known “RGD adhesive loop”, which plays a very important role in the
interaction with specific integrin receptors (Figure 13) [90]. They are classified in
four different groups depending of their polypeptide length, number of subunits,
and the number of disulfide bonds each has. They work as anti-adhesive, anti-
migratory and anti-invasive molecules by binding and interfering with integrins.

Contortrostatin (CN) a homodimer [101] and vicrostatin (VCN) a monomer
[102] are two disintegrins that contain an RGD tripeptide motif that binds with
high affinity to integrins. Contortrastin was originally isolated from the venom of
the southern copperhead snake. CN is a cysteine-rich protein (10 cysteines per
monomer) displaying almost no secondary structure and a complex folding pattern
that relies on multiple disulfide bonds to stebilize its tertiary structure [103]. It
binds to integrins of the β1, β3 and β5 subclasses, including receptors for
50
fibronectin (α5β1), vitronectin (αvβ3, αvβ5), and fibrinogen (αIIbβ3) [104]. The
major issue with CN is that isolating it from the copperhead venom produces very
low amounts due to the low amount of CN in the venom plus the amount loss
during purification. This is why a recombinant molecule with the same biological
activity than CN was developed and called Vicrostatin. Vicrostatin (VCN) is derived
from CN and was produced in the research laboratory using recombinant DNA
technology.

2.1.5 Silicone
Silicones are polymeric materials that have been widely used in medicine
for their biocompatibility and their unique chemical and physical properties.
Silicones maintain their flexibility thorough a wide temperature range, their
chemical stability enables biocompatibility and biodurability in long-term
applications, they show no chemical interaction with foreign molecules, they are
fully transparent to visible and infra-red light which it gives them an advantage
over other materials for imaging, and they absorb light photons in the UV range,
typically at and below 280 nm wavelength (at an above 4.4 eV photon energy)
[105], [106].

The word “silicone” refers to a general and broader category of synthetic
polymers whose backbone is made of repeating silicon-to-oxygen bonds. They are
also bonded to organic groups, which are usually methyl groups. In the average
51
cases there is one silicon atom for one oxygen and two methyl groups [105].
(Figure 14)

Figure 14. Silicone structure. Silicones consist of an inorganic silicon (Si) – oxygen
(O) backbone chain with organic side groups (R) attached to the silicon atoms. The
organic groups are usually methyl groups (example: !"
!
).

To be able to bond foreign species to silicone, it is necessary to “open” their
structure by modifying its atom assembly and breaking some of the inter-atomic
bonds. This process cannot be accomplished by mechanical force or thermal
changes because of the silicone elasticity and its flexibility through a wide range of
temperatures [107]. Laser irradiation results in selective surface decomposition of
silicone, preserving the inorganic silicon-to-oxygen backbone structure of the
polymer and eliminating the organic part. Using UV laser source at 248 nm limits
photon absorption to Si-R bonds, which it leaves a polymeric chain that is formed of
Silicon-to-Oxygen monomers. Both the irradiated surface and the lateral debris are
electrically “active”, being negative charged due to the two dangling bond electrons
that are attached to each silicon atom [104]. For our project purposes, these two
Silicon- dangling bond electrons can be utilized in attaching CN and VCN covalently
to the surface of the silicone.
52
2.2 Specific Aim
The goal of this study is to use the disintegrin technology as a new novel
method of attachment for the retinal prosthesis utilizing a silicone modified with
bioactive materials.

For the integrin–disintegrin approach to work, integrins must be present in
the target tissue (retina). Integrin subunits a4, a5 and b5 were present in the inner
limiting membrane (ILM) of the retina in rabbits and dogs (Figure 15) proving a rich
substrate for integrin-disintegrin binding. Previous experiments have shown that
VCN and CN bind to these integrins [106] [108].

Figure 15. Rabbit retina integrins. (A) Panel on left demonstrates !5 integrin (yellow
arrow, red fluorescence) in the ILM. (B) Panel in the middle shows absence of !3. (C)
Panel on right demonstrates !5 integrin presence (white arrow, green fluorescence)
in the ILM of the rabbit retina. Blue fluorescence is DAPI counter staining of the nuclei
of the neurons. Grey images are photomicrographs. Lower right image of each panel
is overlay of all three images. Image taken from [104].

53
2.3 Material and Methods
2.3.1 Silicone- Disintegrin Process
Experiments on selective surface irradiation were performed on several
commercially available formulations of silicone to investigate whether an active
area could be created. These experiments studied both the physical ablation/
debris field area on the surface, and the degree of disintegrin binding to the
modified surface area under different laser processing conditions (patterns, depth,
and size of the lased area). NuSil MED-4800 family (4810, 4830, 4840, 4850, 4860)
and other silicones (NuSil MED 4286a DowCor WL5150) were tested.

To modify the silicone surface, an ATL ProMaster Excimer Laser was used
with a 248 nm wavelength under KrF medium. The laser beam was focused on the
sample through a given geometrical shape. Table 1 shows the settings that showed a
consistent debris field area on the surface, even depth and clean edges, as judged
by visual inspection. These parameters were used during the rest of experiments.

54
ATL ProMaster Excimer Laser Settings
Pattern: Ring Structure
Electron beam power 10KeV≤
Pressure 6500≤
Pulse rate 150 Hz
Geometrical Shape Circle
Beam size (Circle) 250 !"
Space between circles 250 !"
Number of turns 5
Overlap 5
Repeat Rate 50 Hz
Table 1. An ATL Promaster Excimer Laser was used during all the experiments
performed. Shown in this table are settings that showed a consistent debris field area
on the silicone surface, thus were used during all the experiments reported here.

CN and VCN were evaluated for their binding affinity to the active silicone
surfaces and for their ability to support cellular growth. After lasing, silicone
samples were exposed to the disintegrin solution, incubated overnight, washed in
phosphate buffered saline (PBS), and separated into two groups. In Group 1,
attachment strength to retina was measured in vitro to determine if the
disintegrins were tightly bound to the laser modified silicone. In Group 2, ovarian
cancer cell line OVCAR-3 was plated on the sample and images recorded every 6
hours (up to 48 hours post incubation) to measure the ability of the disintegrins to
55
promote cellular attachment and growth. The cells were transfected with green
fluorescent protein (GFP) to aid visualization.

2.3.2 Silicone Sterilization
To consider a device for human use, a sterile assurance level is required. The
main issue with the most common FDA- approved terminal sterilization techniques
are not bioactive-material-friendly, causing chemical modifications on biological
materials [109]

During our experiments we used ISO 11135 Ethylene Oxide (EtO), which it
is approved for medical devices. To make sure EtO was bioactive material-friendly,
we tested two different protocols to verify we were not causing any modifications
of the biological materials:
1.) Silicones were sterilized with EtO before lasing process. After sterilization, the
lasing process was performed in a clean environment, followed by exposure to
disintegrin solution.
2.) Silicones were lased, exposed to the disintegrin solution overnight and then
sterilized with EtO.

Both methods were tested by evaluating disintegrin- Retina attachment in-
vitro. Experiments described below.

56
2.3.3 In-vitro Experiments
Cadaveric pig eyes (Sierra Medical Supply,Inc.) were used for in-vitro
experiments. The mixed-breed pigs were all healthy and aged between 5-7 months.
The time between animal sacrifice and the start point of our experiment was less
than twelve hours. Diluted (Dulbecco’s PBS) plasmin (EMB Biosciences, Inc)
solution (0.15 ml) was injected into the vitreous cavity 3 days before VCN-silicones
were implanted, to create a posterior vitreous detachment (PVD) and facilitate
removal of the vitreous. During the three-day period, eyes were kept refrigerated
at 4 celsius. The eye was placed on a foam board and pinned to the board to keep it
in place. Three incisions, similar to standard vitrectomy, were made in the sclera to
place the infusion line, the vitrectomy cutter and the light 20 gauge instrument
(Stellaris 20 g vitrectomy pack) were used. The vitreous was then completely
aspirated (Stellaris, Bausch & Lomb). After vitrectomy was completed, the cornea
of the pig eye was removed to facilitate epiretinal placement of the VCN-silicones
samples. Samples were held in place for 20-30 seconds, and then pulled off using
forceps. The surgeon did not have knowledge of what type of silicone or disintegrin
was used. The surgeon then estimated the relative strength of attachment in each
case.

2.3.4 Surgical Procedure – In vivo Experiments
Adult pigmented rabbits (Irish Farm), ~3 months old were used for all
experiments. Implantation of the VCN-silicone was performed in the left eye of each
57
animal (n=4). After VCN-silicone implantation, animals were kept for a period of
three months and then euthanized. All animals were maintained on a daily 12 h
light/dark cycle. All procedures were in conformance with the Guide for Care and
Use of Laboratory Animals (National Institutes of Health). The University of
Southern California Institutional Animal Care and Use Committee reviewed and
approved all procedures.

Prior to surgery, 0.15 ml diluted plasmin solution was injected into the
vitreous cavity while the eye was observed under a surgical microscope. This
solution helps liquefy the vitreous and makes it easy to perform vitrectomy without
causing retinal detachment. Thorough removal of the vitreous allows clean
exposure of the epiretinal surface, which has binding sites for disintegrins. A week
after injecting plasmin, the VCN-silicone implantation surgery was performed.

The pupil was dilated with three drops each of 1% tropicamide and 2.5%
phenylephrine. Three incisions were made in the sclera to place the infusion line,
the vitrectomy cutter and the illuminating fiber optic probe. 25 gauge instruments
were used (Stellaris, Bausch & Lomb). The vitreous was then aspirated and
triamcinolone acetonide (Triesence) injectable suspension (40 mg/ml) was
injected during vitrectomy to stain the vitreous and allow the complete removal of
the vitreous and produce a clean epiretinal surface. The VCN-silicone was then
inserted into the vitreous cavity with the disintegrin surface facing the retina; it
was placed on the epiretinal surface, close to the optic nerve. Care was taken to
58
avoid contact between VCN-silicone and blood, as this would contaminate the
disintegrins. Placement was evaluated under a surgical microscope. Once the VCN-
silicone was properly positioned, it was held in place for 20-30 seconds to promote
disintegrin-integrin binding. After implantation, the infusion line and light were
removed and incisions were sutured with 6-0 vicryl.

Post-operative exams were done every two weeks for a three-month period.
If the optical path was clear, OCT images were taken to evaluate placement,
attachment of the VCN-silicone, and damage of the retina. The eye was observed for
signs of endopthalmitis, to test the sterilization method. After three months
animals were euthanized and eye was enucleated for histology analysis.

2.4 Results
2.4.1 Silicone - Disintegrin process
Six different types of silicone were evaluated to see if an active area could be
created. Silicones were lased and scanning electron micrograph (SEM) images were
taken from each sample (Figure 16) We visually assessed each sample for physical
ablation/debris field area on the surface, the amount of material removed,
definition of edges, evenness, depth, and symmetry of the lased area. Figure 16
shows how MED 4850 (Figure 16E) and MED 4860 (Figure 16F) have edges that
are clean and defined, uniform ablation area, and a large debris field. Other
silicones like MED 4286 (Figure 16A) and WL5150 (Figure 16B) show poor
59
definition on the edges, poor debris field area, and unevenness in the lased area.
MED 4830 (Figure 16C) and MED 4840 (Figure 16D) show clean and defined edges,
however, although the ablation area is even it lacks depth and only a small debris
field is observed.

Figure 16. Different types of silicones were lased to evaluate if an active area could
be created. SEM images were taken to each type of silicone after lasing process.
Evaluation consisted on visually assessing each sample for physical ablation/debris
field area on the surface, the amount of material removed, definition of edges,
evenness, depth, and symmetry of the lased area. The silicones shown in this image
are A) MED 4286 B) WL5150 C) MED4830 D) MED4840 E) MED4850 F) MED4860.
MED 4850 and MED 4860 have edges that are clean and defined, uniform ablation
area, and a large debris field.

Twelve more samples (2 per type of silicone mentioned above) were lased
and treated with disintegrins to evaluate the degree of disintegrin binding to the
60
modified surface area (Figure 17). By visual inspection, each sample was evaluated
for growth of OVCAR3, human ovarian cancer cells that display integrins avb3 and
avb5; increased fluorescence indicates more cell growth and a better substrate for
integrin-disintegrin binding. MED 4860 was imaged six hours post plating (Figure
17A) and 48 hours after incubation (Figure 17B), showing a difference in cell
growth between the two time points. At 48 hours we can observe how cells grew
only on the putative location of disintegrins (as defined by lasing). Unlased, control
silicone showed no cell growth 48 hours after incubation (images not shown). In
MED 4830 (Figure 17C), and MED 4840 (Figure 17D) some cell growth was
observed 48 hours after incubation but MED 4850 (Figure 17E) and MED 4860
(Figure 17F) had a confluent layer of cells at the same period of time. MED 4860
was superior in supporting biological function of the disintegrins, as indicated by
the highest level of fluorescence. MED 4286, MED 4810, WL5150 did not show cell
growth during the 48-hour period (images not shown). Table 2 shows a summary of
the results obtained in this experiment.

61

Figure 17. Images taken to evaluate disintegrin binding and cell growth on silicones
after lasing process and disintegrin exposure: A) MED 4860 six hours after plating
cells and B) MED 4860 48 hours after plating cells shows cell growth by light
microscopy on the lased area but not on the unlased areas that were also exposed to
disintegrin. C) MED4830 exposed to OVCAR-3, D) MED4840, E) MED4850, and F)
MED4860. C-F were exposed to OVCAR -3 and cell growth is evaluated by
fluorescence. MED 4830 to MED 4860 support cell growth but MED 4860 is superior
in supporting biological function of the disintegrins (increase in fluorescence
observed due to increased OVCAR3 cell growth).

62
Cell Growth/ Disintegrin Binding
Testing
CN VCN
DowCor WL5150 ---- ----
Nusil MED 4286 ---- ----
Nusil MED 4810 ---- ----
Nusil MED 4830 + +
Nusil MED 4840 ++ ++
Nusil MED 4850 ++ ++
Nusil MED 4860 +++ +++
Table 2. Seven different types of silicones were lased and treated with disintegrins to
evaluate the degree of disintegrin binding. Samples were evaluated for growth of
OVCAR3 by visual inspection. Increase in fluorescence means VCN-silicone supports
cell growth which it is proportional to disintegrin binding.
---- indicates no observable cell growth; + through +++ indicates increasing levels of
OVCAR-3 cells growth (increase in green fluorescence).

2.4.2 In-vitro Experiments
In-vitro experiments were performed to evaluate attachment. Four types of
silicone were used, based on the results above: MED 4830, 4840, 4850 and 4860.
Eight (two of each type of silicone) samples were lased and immediately placed in
disintegrin solution (CN or VCN), incubated overnight, and washed in phosphate
buffer saline (PBS) before being implanted in a cadaveric pig’s eye. After vitrectomy
and corneal removal, samples were given to a retina surgeon; one sample per eye
was used. The surgeon held silicone in place for 20-30 seconds to promote
adhesion, then pulled it off with forceps and reported the observed strength of
attachment.

63
Table 3 shows a summary of the results obtained in this experiment. MED
4830 showed no observable attachment, MED 4840 showed very poor attachment
and it was pulled off very easily from the retina, MED 4850 showed a slightly better
attachment than MED 4840 but it could be pulled off the retina with moderate
force, MED 4860 showed a strong attachment to the retina, and when surgeon tried
to remove the sample the retina stayed attached to the sample and detached from
the eye. There was no observable difference between CN-silicones and VCN-
silicones.
Disintegrin-Retina Attachment In-
vitro
CN VCN
Nusil MED 4830 ---- ----
Nusil MED 4840 + +
Nusil MED 4850 ++ ++
Nusil MED 4860 +++ +++
Table 3. In-vitro experiments in cadaveric pig eyes were performed to study
disintegrin- retina attachment. Four different types of silicones and two different
types of disintegrins were used during these experiments. Attachment of silicone
samples with two different disintegrins. ---- indicates no observable binding; +
through +++ indicate increasing levels of binding.

2.4.3 Sterilization Protocol
For sterilization testing, we continued testing the last four different types of
silicones that showed disintegrin binding (MED 4830, MED 4840, MED 4850 and
MED 4860). Twenty-eight (seven of each type of silicone) samples were used to
64
evaluate sterilization. Samples were divided in four different groups: Group 1 -
Eight samples (2 of each type of silicone) were lased and immediately placed in
disintegrin solution (1 in CN and 1 in VCN), incubated overnight, and washed in
phosphate buffered saline (PBS). Group2 - Eight samples (2 of each type) were
sterilized using clinical ethylene oxide (EtO) protocol, and then lasing was
performed. Group 3 - Eight samples (2 of each type) were lased, placed in
disintegrin solution, incubated overnight, washed in PBS, and then sterilized using
EtO. Group 4. Four samples (one of each type) served as control with no
sterilization or lasing. Samples were washed in PBS before implantation.

Samples were tested in cadaveric pig eyes. After vitrectomy was completed,
and the cornea removed for easy placement, samples were given to the surgeon;
one sample per eye was used (n=28). The surgeon held silicone in place for 20-30
seconds and then removed it with forceps and reported his assessment of the
strength of attachment. Table 4 shows a summary of the results obtained during this
experiment. Non sterile samples (Group 1) and samples lased after sterilization
(Group 2) showed the same results reported above where MED 4830 shows no
attachment to the retina, MED 4840 and MED 4850 shows poor attachment to the
retina and with MED 4860 a strong attachment to the retina was observed. No
difference between CN- Silicone and VCN-Silicone was reported. Samples sterilized
after lasing procedure (Group 3) and control group (Group 4) showed no
attachment to the retina for any of the silicone groups and no difference between
CN-Silicone and VCN-Silicone.
65

Sterilization Protocols / Disintegrin-Retina Attachment
NonSterile
(Group 1)
Control
(Group 4)
Sterilization
before lasing
(Group 2)
Sterilization after
lasing (Group 3)
CN VCN PBS CN VCN CN VCN
MED 4830 ---- ---- ---- ---- ---- ---- ----
MED 4840 + + ---- + + ---- ----
MED 4850 ++ ++ ---- ++ ++ ---- ----
MED 4860 +++ +++ ---- +++ +++ ---- ----
Table 4. Different sterilization protocols were tested to ensure viability of the
disintegrin after sterilization. Group 1 went through the lasing process without any
sterilization method. Groug 2 was sterilized before the lasing process. Group 3 was
sterilized after the lasing process and Group 4 is a control group, silicones were no
lased and placed in PBS. ---- indicates no observable binding; + through +++ indicate
increasing levels of binding.

Based in the results reported above MED 4860 and the group 2 sterilization
method were used for in-vivo experiments. VCN- Silicone was selected because no
difference between CN-Silicone and VCN- Silicone was reported in any of the
experiments and because VCN is a recombinant disintegrin that is more easily
obtainable and the likely path towards translation to a medical device.

2.4.4 In-vivo Experiments
Adult pigmented rabbits, ~3 months old were used for all experiments
(n=4). Two out of the four rabbits implanted showed attachment (Rabbit 1 and 4)
of the silicone to the surface of the retina during the entire evaluation period.
66
Rabbit 2 and 3 did not show silicone-retina attachment but they were evaluated for
feasibility of the sterilization method, since we left the VCN-silicone sample in the
vitreous cavity. Rabbits 1, 2 and 3 were observed for 2 months and rabbit 4 was
observed for three months.

None of the rabbits developed any type of eye infection during the
evaluation period. Rabbit 1 was evaluated every two weeks for a two-month period
with fundus images. VCN-silicone sample was placed very far from the optic nerve,
so OCT images could not be taken. Rabbit 1 showed attachment of the silicone to
the retina surface in every evaluation during the two-month period. Rabbit 4 was
evaluated every two weeks for a three-month period using fundus images and OCT
images (Figure 18). At two weeks after implantation (Figure18-1) VCN-Silicone
attachment to the retina in OCT image was observed (Figure18-1B); fundus image
(Figure18-1A) shows attachment of the silicone to the retina and a small cataract
developing caused during surgery. Fluorescence Angiography (FA) image
(Figure18-1C) shows well preserved vasculature and no evidence of leakage or
pooling of blood. At five weeks after implantation (Figure18-3) VCN-Silicone
attachment to the retina in OCT image was observed (Figure18-3B) and a small
tissue layer was observed growing on top of the VCN-Silicone sample (arrows);
fundus image (Figure18-3A) shows attachment of the silicone to the retina and
cataract stable. FA image (Figure18-3C) shows no vessel leakage. At 3 months after
implantation (Figure18-5) VCN-Silicone attachment to the retina in OCT image was
observed (Figure18-5B), and tissue layer on top of the VCN-Silicone can still be
67
observed; fundus image (Figure18-5A) shows attachment of the silicone and no
progression of cataract. FA image (Figure18-5C) shows no vessel leakage. No signs
of infection was observed throughout implant period. Summary of results are
reported in Figure 19.

68

Figure 18. Images were taken to evaluate disintegrin binding in silicones after lasing
process, retina-silicone attachment and feasibility of sterilization method. During
each time point, three images were taken. Column A shows a fundus image of the
retina with the silicone sample attached to it. Column B shows an OCT image. This is a
cross section of the retina where the sample is attached and Column C shows a
fluorescein Angiography image to evaluate retinal cirulation. Images were taken at 5
different time periods: Row 1) One week after implantation, Row 2) two weeks after
implantation, Row 3) 5 weeks after implantation, Row 4) 2 months after implantation
and Row 5) 3 months after implantation.

After a two and a three-month period for rabbit 1 and 2 respectively,
animals were anesthetized and plasmin was injected in the vitreous cavity near the
sample to detach sample from retina surface. Plasmin allowed us to detach silicone
sample from the retina without causing any damage to the retina. When plasmin
was injected, one side of the silicone sample lifted up from the retina but the other
side stayed close to the retina, due to fibrosis growing on top of the sample. The
69
sample was removed when it was visibly detached from the retinal surface. Retinal
detachment was not observed after sample removal. The eye was then enucleated
and the tissue fixed for processing. Staining with H&E shows no significant damage
to the retina in the area where sample was placed. Histology shows some red blood
cells on the surface of the retina that could have been caused by the surgeon when
removing the fibrosis to release the sample. (Figure 20)

Figure 19. In-vivo experiment. Attachment of sterile silicone samples with
disintegrins. – indicates no observable attachment or infection; + indicates
attachment or infection; x indicates data point was not acquired. At, attachment, In,
infection. 1week (1w), 2weeks (2w), 5 weeks (5w), 2months (2m), 3months (3m)
indicates the data points where rabbits were evaluated. Rabbit 1(R1) and Rabbit
2(R2) showed attachment during the period they were observed and no signs of
infection during the same period. Rabbit 3(R3) and Rabbit (R4) showed neither
attachment nor sign of infection during the period they were maintained.
70

Figure 20. Histology images taken from rabbit 2. (A) Arrows show area where
sample was placed. (B) Zoomed image of area between arrows in image A. No
significant damage is recorded in these images. Image B shows some red blood cells
on the retinal surface (arrows point to the red blood cells).

2.5 Discussion and Conclusion
In this chapter we have described a technique that allows us to use
disintegrin-integrin binding as a novel attachment method for retinal tissue.
Integrins are α/β heterodimeric glycoproteins that are expressed at the surface of
mammalian cells [92] and are involved in the regulation of cell growth and survival
[91]. So far 18 α- and 8 β-subunits have been described for integrins in human
cells; they mediate a wide range of cellular functions, including adhesion [93].
71
Disintegrins are disulfide-rich, RGD-containing peptides that bind to integrins on
cells. CN, a disintegrin originally isolated from southern copperhead venom binds
to integrins, including receptors for fibronectin (α5β1), vitronectin (αvβ3, αvβ5),
and fibrinogen (αIIbβ3) [99]. For the integrin–disintegrin approach to work,
integrins must be present in the target issue (retina). Integrin subunits a4, a5 and
b5 were present in the inner limiting membrane (ILM) of the retina in rabbits and
dogs proving a rich substrate for integrin-disintegrin binding.

We have described a technique to successfully lase and produce an active
area on the silicone surface that will allow disintegrin binding. Our study evaluated
different types of medical grade, commercially available silicones and how they
interact with laser processing. Significant differences were noted based on the type
of silicone used. These differences affected the silicone-VCN binding and the
strength of the silicone-VCN-retina attachment. This knowledge will be useful for
choosing a stronger or weaker attachment, which could be used to control silicone-
VCN-retina binding depending on the particular needs of the device. Future
experiments should be directed towards understanding the fundamental physical
mechanisms that determine binding strength.

Different approaches have been utilized to attach the retina prosthesis to
the retina as an alternative to a retinal tack. To test magnetic fixation, researchers
implanted one magnet in the suprachoroidal space, and a second magnet on the
back of an intraocular component. This arrangement was used to successfully
72
position a retinal prosthesis epiretinally for a 6 week period, maintaining the
position reliably, but causing trauma to the retina on the edges of the device [110].
Other adhesive materials such as fibrin glues, photocurable glues and hydrogels
have been tested, proving hydrogels were much more adherent to the retina than
the other bioadhesives [111]. Mucoadhesive materials such as chitosan, thiomers,
boronate-containing polymers, liposome-based mucoadhesive formulations and
acrylic-based polymers have shown potential for tissue adhesion [112]. Further
studies will have to be performed to evaluate their applicability for specific
applications like retinal prosthesis adhesion.

To consider a device for human use, sterility is an absolute requirement. The
main issue with the most common FDA-approved terminal sterilization techniques
is that they are not bioactive-material-friendly, causing chemical modifications of
biological materials [109]. During our study a sterilization protocol was tested and
proven to be reliable for bioactive materials. This involved sterilization of the
silicone device before lasing using a clinical ethylene oxide protocol and then doing
the lasing and disintegrin attachment under sterile conditions. Other protocols like
supercriticial CO2 (scCO2) in the presence of H2O2 have been studied and have
proven to be reliable for sterilization of bioactive materials [109].

Silicone has excellent mechanical and biocompatibility properties, which
makes it an excellent substrate for interfacing with soft tissue like the retina. This
study demonstrated a technical approach for functionalizing the surface of silicone
73
to promote adhesion, a capability that could be very advantageous in future
medical devices. This process was shown in the retina, but may be applicable to
brain cortex, spinal cord and different areas of the body where sutures are not an
option. Further, this procedure could be adapted for drug delivery in future
modifications, an area that we are already actively pursuing.

74

CHAPTER 3
High Charge Density Stimulation safety
3.1 Background
3.1.1 Neural stimulation
All neurons have 4 functions: Reception, triggering, signaling and secretion.
All signals are received at the dendrites of the neuron, and travel to meet at the
soma (axon hillock) where action potentials are generated if a certain threshold is
reached. Action potentials propagate over the axon of the neuron traveling long
distances without losing amplitude and reaching the synapse point where the axon
meets another neuron and releases neurotransmitters. Neurotransmitters are
received by the receptor, forming a unidirectional connection (Figure 21).
Information travels from neuron to neuron either by electrical or chemical signals.
Electrical signals are produced by rapid changes between the electric current
inside and outside of the cell, driving the electrical potential along the cell
membrane away from its resting state [2].
75

Figure 21. Neuron Schematic. Neurons are excitable cells that function to process
and transmit information. Information travels from neuron to neuron either by
electrical or chemical signals.

Neural stimulation uses the electric component of the nervous system by
locally altering the membrane voltage of the neurons either intracelullary or
extracellulary. The main effect of external electrical stimulation is the change of
transmembrane voltage, causing either a depolarization or a hyperpolarization by
manipulating the kinetics of the voltage-gated channels and having an action
potential as a response [113].

3.1.2 The electrode- Tissue interface
To create action potentials with an external stimulation, an electrode needs
to be brought close to the target neural structure. Within the implant itself,
electrical current is carried by electrons, but within tissue, current is carried by the
76
movement of charged ions, this is why at the electrode – tissue interface some
interactions have to take place between the electrons in the electrodes and the ions
in the tissue to maintain electrochemical equilibrium also known as electrical
double layer. The electrical double layer could be formed for charge accumulation
or an electrochemical reaction [114].

Charge accumulation at the interface is caused by ions in the electrolyte
combining with the electrode, which leads to a net transfer of electrons between
the two phases causing a plane of charge at the metal electrode that is opposed by a
plane of charge in the electrolyte. Electrochemical reaction is caused by adsorption
of certain chemical species and preferential orientation of polar molecules such as
water [115].

Both charge accumulation and electrochemical reaction can be
characterized as reversible and irreversible processes. Reversible currents can be
due to charge accumulation (capacitive currents), but also due to reversible
electrochemical reactions (pseudo-capacitive currents), while irreversible currents
are due to electrochemical processes that cannot be reversed such as oxygen
evolution. This is important because for any neural excitation to take place, current
has to flow through tissue. The dynamic interaction between electrons in the
electrodes and the ions in the body can greatly affect the performance of the device
[116].

77
The most favorable operation of the electrodes occurs when charges move
across the interface reversibly and do not produce toxic products that diffuse in to
the electrolyte or degrade the electrode.

3.1.3 Equivalent circuit model
The device-tissue interface has nonlinear impedance that is commonly
reduced to a parallel resistor-capacitor (RC) circuit model (Figure 22). The resistive
element (!
!"#"$"%&
) represents faradaic charge transfer where electrons are
actually passed between the electrode and the tissue during chemical reactions like
reduction and oxidation. The capacitive element (!
!"
) models non-faradaic charge
transfer where no charge carriers cross the interface and currents are generated in
the tissue using electromotive force to redistribute charged species [117].

Figure 22. (A) Device-Tissue interface. (B) A dynamic model at the device-tissue
interface when a current is applied at the electrode.
78
3.1.4 Stimulation Characteristics
Neural stimulation can be implemented in different ways: cathodic vs anodic
stimulation, monopolar vs bipolar stimulation, and current vs voltage control. Also
it is important to ensure proper function during the lifetime of the device, this is
why materials like titanium, platinum, or iridium are used to make the electrodes.

3.1.4.1 Cathodic and anodic stimulation
When an electrode delivers charge, some redox reactions occur at the
electrode surface leaving some products (!
!
!" !
!
) that if it propagates away from
the surface can cause damage to the electrode and the tissue. To avoid this, systems
use charge balanced biphasic pulses to stimulate. The first phase injects charge in
order to evoke a response in the target cell, followed by an equally in charge but
opposite phase, which removes charge and reverses electrochemical processes.

Different schemes of stimulation can be used called cathodic first or
cathodic excitation and anodic first or anodic excitation. During cathodic excitation,
the cathodic current will propagate in each direction from the excitation node. The
electric field pattern creates small hyperpolarizations lateral to the area of
depolarization that does not affect the action potential propagation. During anodic
excitation, the anodic current is so large creating a depolarization on either side of
the hyperpolarization and generating an action potential that will propagate in
opposite direction than the action potential created by a cathodic stimulation.
79
Thresholds of anodic excitation are 3-7 times higher than the thresholds of
cathodic excitation and this is why cathodic first pulses are commonly used during
neural stimulation [118] .

3.1.4.2 Monopolar and bipolar stimulation
To inject current an active electrode and a return electrode are required.
Current will flow from the active electrode to the return electrode. When both
electrodes are placed far away so the current radiates out of the active electrode is
called a monopolar configuration (Figure 23). During monopolar configuration the
current injected by the active electrode is distributed evenly in all directions and it
could be used to increase the stimulation distance [119].

When the return electrode is placed so close to the active electrode that the
current pathway is concentrated between the two electrodes leading to a localized
stimulation is called bipolar configuration (Figure 23). During bipolar configuration
threshold current is reduced when the return electrode gets closer to the active
electrode, which it leads to localized stimulation in the tissue immediately
surrounding the electrodes.

80

Figure 23. Monopolar and bipolar stimulation configuration. Current is distributed
evenly in all directions with a monopolar configuration. Current is concentrated
between the two electrodes leading to a localized stimulation with a bipolar
configuration.

3.1.4.3 Voltage controlled vs. Current Controlled stimulation
Voltage controlled systems depend of the tissue resistance to limit the
current. The voltage drop across the tissue decreases over time due to voltage that
builds up across the capacitor and the electrode interface, making in it less safe due
to the lack of control over the injected charge into the tissue. One of the advantages
of voltage-controlled systems is that it offers high power efficiency, and the
possibility to stimulate different neuron populations that responds to different
pulse configurations due to the initial current spike.

Current-controlled configuration is safer because the output voltage is
increased to maintain constant current. Since the current injected into the tissue
doesn’t depend of the tissue resistivity or distance between the electrodes and the
tissue, current controlled configurations are a better option when planning on
81
using a multi electrode matrix. One of the disadvantages of using current-
controlled systems is the increased complexity in the circuit design [120].

3.1.5 Stimulation Electrode
For the development of neural stimulation system, electrodes need to be
able to handle harsh environments, have lifetime durability, inject large amounts of
charge, long-term electrical and mechanical stability and be biocompatible.

Low impedance electrodes are more efficient, since less energy is required
for current to be passed to tissue. For neural activation selectivity smaller
electrodes are ideal. Small electrodes will allow targeting cells without activating
the adjacent cells. The issue with small electrodes is that the impedance is higher,
reducing the effectiveness of the electrode. Since increasing the geometric area is
not an option, deposition processes and selection of proper materials are options to
increase the surface area. Materials like titanium, platinum and iridium have shown
to be well tolerated by the tissue without concerns related to toxicity.

Platinum is a widely used material because of its durability and its relative
ease of manufacture. It is resistant to corrosion and can transfer charge using
reversible reactions that do not harm the tissue. Platinum can safely supply 0.1 to
0.4 0.4!" !"
!
of charge. Iridium can inject higher amounts of charge (1!" !"
!
)
than platinum through oxidation reactions for the same electrode area. Platinum’s
82
high mechanical strength leads to difficulties in machining or molding, and this is
why it is usually sputtered or electroplated onto the electrode surface, adding
complexity to the manufacturing process. Titanium supplies low amounts of charge
within the operating voltage range, and this is why it is often used as a casing for
the electronics [121].

DC reactive magnetron sputtering [122], DC reactive sputtering [123] are
methods used to increase the surface area of electrodes but they are not very
efficient or cost effective because it requires high vacuum conditions, large target
size and low material deposition efficiency. In contrast, electrochemical depositions
can be performed at room temperature and under atmospheric pressure, which it
makes it a more affordable method.

Platinum-Iridium alloys have been widely during biomedical applications,
taking advantage of the properties of both Materials. During our study, we will be
using a custom made platinum/Iridium electrode (80%/20%) coated with a
deposition method developed at our laboratory [124] with Platinum/Iridium at the
tip (60%/40%) to increase the roughness of the electrode, allowing us to test high
charge densities safely.

83
3.1.6 Safety stimulation
Based on everything mentioned above, it is important to pick the right
parameters to have a safe neural stimulator system for both the tissue and the
electrode itself. Parameters like electrode size, electrode shape, material,
stimulation parameters and placement need to be chosen properly to reassure
safety.

3.1.6.1 Causes of electrical damage
Damage of neural tissue during electrical stimulation could be caused by
different factors like example for mechanical force, for tissue overstimulation or for
electrochemical processes.

As mentioned in our chapter 2, the distance between the electrode and the
tissue is very important to maintain low thresholds during electrical stimulation,
but a study done by Colodetti states than an unquantified amount of mechanical
pressure from a probe held in contact with a rat’s retina for less than an hour
caused significant disruption of the retina layers [125].

Intrinsic biological processes cause tissue damage when the tissue is
overstimulated. Over stimulation of the tissue can cause neuron firing for long
periods of time, changing ion concentration intracellular and extracellular,
84
depletion of oxygen and glucose, and excessive release of glutamate, creating a
toxic environment for the neuron. [126].

Retinal damage can also be created due to toxic electrochemical reactions
formed during electrical stimulation higher than that the physiological system
could tolerate or/and when the safety limits of the electrode are exceeded.

Another possible factor for neural damage is electroporation [127].
Electroporation is caused by large depolarizations and hyperpolarizations creating
cellular damage due to dielectric breakdown of the bilipid membrane and can also
cause damage to the voltage-dependent gating mechanisms of ion channels.

3.1.7 Methods for visualizing damage
3.1.7.1 Hematoxylin and Eosin Staining
Hematoxylin and Eosin stain is commonly used to study morphological
changes in various tissue types. This type of staining has been used for centuries
with minimum changes. Hematoxylin is a basic dye that stains DNA in the nucleus,
and RNA in ribosomes with a purple color. Eosin is an acidic dye that stains
proteins like cytoplasmic filaments, intracellular membranes and extracellular
fibers with a pink color [128].

A limitation of hematoxylin staining is that it is incompatible with
immunofluorescence. When immunofluorescence is needed the most common
85
procedure is to leave a serial of sections between H&E stainings to perform
immunofluorescence.

3.1.7.2 Optical Coherence Tomography
Optical coherence tomography (OCT) images detailed ocular structures
noninvasively in vivo with high-resolution images making it an essential tool in the
ophthalmology world. A high-bandwidth light beam is directed on to the target
tissue and the scattered back-reflected light is combined with a second beam
(reference beam), which was split off from the original light beam. The resulting
interference patterns are used to reconstruct an axial A-scan, which represents the
scattering properties of the tissue along the beam path. Moving the beam of light
along the tissue in a line results in a compilation of A-scans, a two dimensional
cross-sectional image of the target tissue can be reconstructed and this is called a
B-scan. Spectral Domain OCTs acquires A-scans by using an array of detectors
instead of using multiple reference beams from a moving mirror.

The most commonly used parameter used from OCT images is retinal
thickness, which is very useful for identifying changes from the normal anatomy
and also changes overtime [129].

86
3.2 Previous research
Different studies have been done to evaluate the safety of electrical
stimulation in neural tissue and in the retina.

Shanon [130] presents a model of safe levels of electrical stimulation based
on data acquired during animal experiments with surface electrodes in cortical
neurons. Figure 24 summarizes the results showing the charge and charge density
at which damage occurs. They found a boundary between safe an unsafe charge
injections that is represented by the equation: log ! = !− log(!) where D is
charge density in !"#$%#&'(/!"
!
/!ℎ!"# and ! is charge in !"#$%#&'(/!ℎ!"#.
When k=2 is the limit of safe stimulation. For different values of k it can be found if
the equation lays in the safety limits or not. This model presented a great
framework for designing future animal safety studies, but now we know that other
factors like pulse duration, pulse rate, stimulus duty cycle and duration of exposure
need to be taken into account when safety stimulation is being evaluated.

87

Figure 24. The hatched area represents the region of charge and charge density
where neural damage is observed. The solid lines represent the proposed model with
parameters k=1.0, 1.5 and 2.0.

Ray [131] performs an experiment with a Platinum/Iridium electrode
implanted into a rat’s eye close to the retina and held in place by a
manicromanipulator. Impedance was used as a method to sense proximity to the
retina. The retina was stimulated for 1 h using a monopolar electrode configuration
(75 um diameter). During the 1 h stimulation period, cathodic-first, charge-
balanced biphasic pulses were delivered to the stimulatimg electrode. Each phase
of the pulse was 0.5 ms long with a 0.1 ms intraphase delay. 0.68 mC cm-2
corresponds to approximately double the safe charge density of the stimulation
electrode. 0.1mC cm-2 is a best approximation of the current clinical settings. The
88
effects of continuous (1 hour) electrical stimulation were determined by evaluating
the retinal morphology and by measuring the electrically evoked responses in the
superior colliculus. This paper reported that the retina is able to tolerate one-hour
stimulation with charge densities of up to 0.68 mC.cm-2 (above safety limits) and
frequencies of up to 300 Hz showing minor morphological changes in the retinal
histology (Figure 25). Histology was examined 3 days and 14 days after stimulation.
Continuous electrical stimulation caused a threshold increase of the electrical
evoked response in the superior colliculus.

Figure 25. Images of a rat’s retina with hematoxylin staining. H&E staining showed
no morphological changes to any of the experimental conditions evaluated in this
paper. Figure B is a control retina and Figure C is a retina stimulated at a high
frequency. No layer disorganization is observed in the magnification insets.
89

Cohen [132] developed a method to study the effects of electrical
stimulation stimulation in the retina under an epiretinal electrode in real time
using an optical coherence tomography (OCT) and a superfused retinal eyecup
preparation. During stimulation, images were collected every 15 seconds during
the 5 minute of pulse train stimulation, and during the recovery period every 30
seconds for another 20 minutes. They used biphasic cathodic first pulses, 1 ms
pulse width and 50 Hz applied through a transparent pipette (0.38 mm i.d., 0.89
mm o.d.). The charge density used was from 23 to 749 !".!"
!!
.!ℎ
!!
. They
reported that stimulation at 44–133 !".!"
!!
.!ℎ
!!
had little effect on the retina;
however, stimulation ≥ 442 !".!"
!!
.!ℎ
!!
caused increases in the reflectance of
the inner plexiform layer (IPL) and edema (Figure 26). The damage seen in retinal
OCT images matched the pattern observed in histological sections, and in the PI
staining.

Most studies examining the safety of electrical stimulation have relied on
post-stimulation methods like histopathological imaging [115], [133]. This study
evaluated the effects of electrical stimulation in the retina using OCT imaging in an
in-vitro model showing us the value of real-time imaging to analyze the retina
response to electrical stimulation during a time course instead of one time point
after stimulation [132].

90

Figure 26. OCT images at different time points during stimulation. Four different
charge densities were evaluated. Number on top of each quadrant (4 quadrants)
represents the charge density used during that specific experiment. Each quadrant is
composed of three OCT images. From top to bottom, first is an OCT image zero
minutes into the stimulation, middle image is an OCT image right after stimulation
was finished and last image is an OCT image fifteen minutes after stimulation was
over.

3.3 Specific Aim
A high-resolution array will require smaller electrodes and closer proximity
between electrodes. Smaller electrodes will help focusing the stimulus, and it will
help with the formation of round, defined, focal percepts which are needed for
91
complex perception, but it also requires higher charge densities than larger
electrodes for the same amount of charge (Figure 27). It is not known if these new
charge density values are safe to stimulate the retina, thus creating a safety
concern.

Figure 27. Charge density for different electrode sizes. Injecting the same amount of
charge through three different size electrodes. The smallest electrode requires the
highest charge density.

We developed an in-vivo experimental model that will help study the effects
of high charge density stimulation in the retina in real time using OCT (optical
coherence tomography). Using an in-vivo model allows us to study the retina in its
natural conditions. We believe that our experiments will help us better understand
the retinal response to electrical stimulation under conditions that resemble the
operating conditions of current and future retinal prostheses.

40 nC
520 µm
40 nC
260 µm
40 nC
75 µm
0.02 mC/cm
2
0.07 mC/cm
2

0.91 mC/cm
2

92
3.4 Materials and Methods
3.4.1 Animal Model
All animals were maintained on a daily 12 h light/day cycle prior
experiment. All procedures conformed to the Guide for Care and Use of Laboratory
Animals (National Institute of Health). The University of Southern California/ The
University of Michigan Institutional Animal Care and Use committee reviewed and
approved all procedures.

3.4.1.1 University of Southern California
Adult pigmented rabbits (Irish Farms, Norco, CA); approximately 2 months
old were used for all experiments. Experiments were performed in only one eye of
each animal (n= 38).

3.4.1.2 University of Michigan
Adult Dutch Belted rabbits (Covance inc., Battle Creek, MI); weighting
approximately 2 kilograms were used for all experiments. Survival experiments
were performed in only one eye of each animal and non- survival experiments were
performed in the other eye right before euthanasia (n= 8).

93
3.4.2 Anesthesia
3.4.2.1 University of Southern California
Rabbits were anesthetized with 4 parts ketamine (100mg/kg; KETASET, Fort
Dodge, IA) and 1 part xylazine (20 mg/kg; X-Jet SA, Butler, Dublin, OH) by
intramuscular injection. One full dose was given at the beginning of the experiment
and half doses every 30 minutes until the end of the procedure.

3.4.2.2 University of Michigan
Rabbits were anesthetized with 4 parts ketamine (100mg/kg; KETASET, Fort
Dodge, IA) and 1 part xylazine (20 mg/kg; X-Jet SA, Butler, Dublin, OH) by
intramuscular injection and maintained during the procedure with Ultane
sevoflurane (250 ml bottle, NDC:0074-4456-04; AbbVie Inc., North Chicago, IL)

3.4.3 Surgical Procedures
3.4.3.1 Non-Survival Experiments- University of Southern California
Rabbit’s forehead was shaved after anesthesia for better placement of the
return electrode. A needle electrode was used as a return electrode and it was
placed in the rabbit’s forehead close to the eye used for the procedure. Rabbit was
placed on a custom-made metal board that allowed better placement for imaging.
The eye was dilated with two drops each of 1% tropicamide (Tropicacyl, Akorn
94
Inc., Buffalo Grove, IL, USA) and 2.5% phenylephrine (AK-Dilate, Akorn Inc., Buffalo
Grove, IL). The animal’s rectal temperature, heart rate, blood pressure and number
of breaths per minute were monitored during surgical procedure and recorded
every 15 minutes. Body temperature was regulated and maintained at 37 °! with
an electric heating pad.

A pediatric barraquer eye speculum (Karl Storz SE & co. Tuttlingen,
Germany) , was placed under anesthesia to facilitate placement of the electrode and
imaging. OCT images were taken as control before electrode placement (HRA+ OCT
Spectralis, HEIDELBERG, Germany). An incision was made in the sclera, 3 mm from
the limbus to avoid damaging the lens and puncturing the retina and a 25 gauge
valved trocar (Alcon, Forth Worth, TX) was used to help keep the stimulating
electrode in position (Figure 28). After experiment was finished, animals were
euthanized by an intravenous injection of pentobarbital (30 mg/kg; Butler, Dublin,
OH) and the eye was enuclated for histology.

3.4.3.2 Survival Experiments – University of Michigan
A needle electrode was used as a return electrode and it was placed in the
rabbit’s forehead close to the eye used for the procedure. Rabbit was placed
sideways on a surgical table and intubated during the procedure. The eye was
dilated with two drops each of 1% tropicamide (Tropicacyl, Akorn Inc., Buffalo
Grove, IL, USA) and 2.5% phenylephrine (AK-Dilate, Akorn Inc., Buffalo Grove, IL).
During the procedure one drop of 0.05% hydrocholoride ophthalmic solution
95
(Tetracaine, Basuch + Lomb, Rochester, NY) was used every thirdy minutes as an
eye anesthetic. The animal’s rectal temperature, heart rate, respiration rate, SPO2,
sevo percentage and blood pressure were monitored during surgical procedure
and recorded every 15 minutes. Body temperature was regulated and maintained
at 37 °! with an electric heating pad.

A pediatric barraquer eye speculum (Karl Storz SE & co. Tuttlingen,
Germany) was placed under anesthesia to facilitate placement of the electrode and
imaging. OCT system (Envisu R-Class; Leica Microsystems, Wetzlar, Germany) was
placed in a custom made holder right on top of the rabbit’s eye. Images were taken
as control before electrode placement. A small section of the conjunctiva was
removed and an incision was made in the sclera, 3 mm from the limbus to avoid
damaging the lens and puncturing the retina (Figure 28).

After experiment was finished, animals were recovered and kept for a
period of 24 hours or 2 weeks. OCT images were taken every three days and after
the two weeks animals were euthanized by an intraperitoneal injection of
pentobarbital (30 mg/kg; Butler, Dublin, OH) and the eye was enuclated for
histology.

96

Figure 28. Left Image shows the setup used during the non-survival experiments
performed in The University of Southern California. Right Image shows the setup used
during survival experiments performed in The University of Michigan.

3.4.4 Stimulation Electrode Insertion
The stimulating electrode was held by a three axis translational stage
(Model 4044 M Parker DAEDAL, Cleveland, OH) mounted on a magnetic base-
articulating arm. The stimulating electrode was advanced inside the eye until it
became visible in the fundus view of the OCT system. The scan line was adjusted to
be along the length of the electrode tip and the retina. The micromanipulator was
used to advance the electrode towards the retina for better positioning without
contacting the retina (Figure 29).

After electrode placement, a fundus image and an OCT image were acquired.
During stimulation OCT images were acquired every two minutes. After
97
stimulation, OCT images were acquired every 5 minutes during a 15 min period.
Fluorescein angiography (FA) images were taken by injecting a fluorescent dye into
the blood stream. The purpose of the dye is to highlight the blood vessels in the
retina so they can be photographed. FA images are used to look for any leakage in
the blood vessels caused by the electrical stimulation.

After the experiments, all animals were euthanized by an injection of
pentobarbital (30 mg/kg; Butler, Dublin, OH) and eye was enucleated for histology.

Figure 29. Diagram of the experiment setup. The electrode was placed through the
sclera and held in place with a micromanipulator. Return electrode was placed in the
rabbit’s head. A spectral domain optical coherence tomography (OCT) system was
placed in front of the rabbit’s left eye and scanned the retina surface while the retina
was being stimulated.

98
3.4.5 Stimulation Electrode
The stimulating electrode was a 150 microns concentric monopolar 90%
Pt/10%Ir electrode (Custom made by FHC, Inc., Bouidain, ME) with a flat tip (Figure
30). A platinum needle inserted in the forehead close to the eye used for the
experiment was used as a return electrode.

The stimulating electrode was modified by forming a high surface area
Platinum-Iridium film on the surface of the electrode using an electrodeposition
technique developed by our laboratory [124]. A potential range of E= +0.2 V to -0.2
vs. Ag/AgCl at scan rate of 0.2 mV/s was used for electrodeposition of 60% Pt/40%
Ir films. The solution was agitated using an ultrasonic homogenizer (Misonix, Inc.
Newton, CT, USA) at a frequency of 20kHz to maintain constant mass transfer
during electrodeposition and to maintain the temperature of the plating solution
around 70℃.

Figure 30. Stimulation electrode schematic. Electrode was custom made by FHC, Inc.

99
3.4.6 Experimental Setup
A four-channel MCS system was used to deliver the stimulus pulses to the
retina. Current and voltage waveforms were monitored and recorded constantly
during the duration of the experiment. Waveforms were recorded using the math
mode on an oscilloscope (Model TDS2000C, Tektronix, Beaverton, OR). A 10k-
Ohms sense resistor was used from a custom made circuit board to record the
current and voltage waveforms. Blocking capacitors were used to minimize DC
current (Figure 31).

Figure 31. Diagram of experimental setup. A MCS system delivers the stimulus pulses
to the retina. Working electrode goes in the vitreous cavity and return electrode in
the rabbit’s head. Current and voltage waveforms are monitored with an oscilloscope.

100
3.4.7 Experimental Groups
3.4.7.1 Non Survival Experiments
Charge-balanced, cathodic first, biphasic current pulses were delivered to
the epiretinal surface. Two different pulse widths (1 and 25 ms), three different
charge densities (0.92, 1.22 and 1.63 !".!"
!!
) And four different frequencies (20,
100, 200 and 333 Hz) were tested. Pulse duration was kept constant for
frequencies 333, 200 and 100 delivering 600.000 pulses which it meant a pulse
duration of 30, 45 and 100 minutes respectively (Table 5). For 20 Hz the pulse
duration was 100 min. A combination of 9 different experimental groups was
tested as showed in the table below.

Table 5. Parameters used during stimulation. Experimental groups were based on
different combinations of charge densities, stimulus frequency, pulse widths, and
duration of pulse train stimulations.

101
3.4.7.2 Survival Experiments
From the information acquired during our non-survival experiments, 5
different groups were selected for the survival experiments. Based on the changes
on retinal thickness three groups that showed changes in retinal thickness and two
groups that didn’t were selected.

Charge-balanced, cathodic first, biphasic current pulses were delivered to
the epiretinal surface. Only 1 ms pulses were delivered, three different charge
densities (0.92, 1.22 and 1.63 !".!"
!!
) And three different frequencies (20, 100
and 333 Hz) were tested. Pulse duration was kept constant for frequencies 333
and 100 delivering 600.000 pulses, which it meant pulse duration of 30 and 100
minutes respectively. For 20 Hz the pulse duration was 100 min (Table 6).

The groups that showed no changes in retinal thickness were kept for 24
hours and a second set of OCTs was acquired and eye was enucleated and sent to
histology. The groups that showed changes in retinal thickness were kept for a
period of 2 weeks and OCTs were acquired every three days to evaluate if the
damage was permanent or temporal.
102

Table 6. Parameters used during stimulation and survival experiments. Experimental
groups were based on different combinations of charge densities, stimulus frequency,
pulse widths, and duration of pulse train stimulations. The green represents the
groups that showed no changes in retinal thickness during non-survival experiments,
and the red are the groups that showed changes in retinal thickness during non-
survival experiments.

3.4.8 Data Analysis
3.4.8.1 Imaging analysis
Both OCT systems used during this experiments have their own software to
analyze the images. The Heidelberg Spectralis Eye Explorer software was used to
analyze all images acquired at the University of Southern Califonia, and the calipers
from the bioptigen OCT were used to analyze the images acquired at the University
of Michigan.

103
Fundus images were acquired to visualize electrode placement (Figure 32).
Electrode was placed close to the visual streak and the optic nerve but not on the
visual streak for better imaging

Figure 32. Fundus image of the rabbit’s eye. Electrode was placed on the inferior
temporal region of the eye, close to the visual streak. Left Image is acquired with the
spectrallis OCT system. Right Image is acquired with the Bioptigen OCT system

Distance between the electrode and the retina was measured using the
calipers from the software (Figure 33). The distance was measured every three
images taken (every 12 minutes) and recorded in excel. Per animal one distance is
reported which represents the average between all the distances measured at the
different time points.
104

Figure 33. Top Left image. Fundus image of the rabbit’s eye. The X scan line was
adjusted to be on the tip of the electrode simultaneously while the electrode was
advanced toward the retina; which helped for placement of the electrode close to the
retina. Top Right image. OCT image. Electrode tip was placed close to the epiretinal
surface. Electrode was advanced towards the retina until the tip of the electrode
could be seen on the OCT scan. Distance between the electrode and the retina was
measured using the calipers function of the OCT software (shown in red). Top Images
were acquired with the spectrallis OCT system. Bottom images were acquired with the
Bioptigen OCT system

105

Retinal thickness was measured to assess damage. The Spectralis software
measures the distance between the two red lines as shown in the image below
(Figure 34). The software gives an estimate but the lines can be individually
modified to have a more accurate measurement. The bioptigen OCT system allowed
us to use multiple calipers to measure the retinal thickness throught the retina as
shown in Figure 34. Retinal thickness was measured every three minutes during the
stimulation period and every five minutes for 15 min after stimulation and
recorded in an excel sheet. For statistical analysis only the measurement before
and after stimulation was used.
106

Figure 34. Retinal thickness measurement. Top Left image is a fundus image of the
rabbit’s eye with the X scan line showing the section of the retina shown in the right
image. Two red lines in the top right image are used to measure the retinal thickness.
Software measure distance between the two red lines as shown with the
perpendicular green line in the top right image and it is reported as shown in the
middle image. Top and middle images were acquired with the spectrallis OCT system.
Bottom images were acquired with the bioptigen OCT system. Bottom left image
107
shows the retinal thickness being measured with multiple calipers identified with
different colors each. Bottom right image shows a fundus image where each color
that represents the caliper is shown to represent the section of the retina that is being
measured.

Fluorescent angiography (FA) are taken to assess vessel leakage caused by
retinal stimulation. Fluorescent angiography images were acquired using the FA
settings from the spectralis system (Figure 35) and analyzed by visualization. FA
images were not acquired in rabbits done at Michigan due to lack of access to a
fundus camera and the previous findings of no leakage even when changes in
retinal thickness were observed.

Figure 35. Fluorescent angiography image. Image of a rabbit’s retina after
stimulation.

108
3.4.8.2 Statistical Analysis
Data was analyzed using Matlab (Natick, Massachusetts, US) and Microsoft
Excel. For non survival a non-parametric t-test (signed-rank test) was performed to
assess if the median difference between the retinal thickness before and after
stimulation is zero. Statistical comparisons were done by performing a non-
parametric one-way analysis of variance (ANOVA- Kruskal Wallis) to assess if the
median of difference in retinal thickness is similar across the eight different groups.
A non-parametric post-hoc comparison was carried out using the Wilcoxon test.
Ranks were assigned to the data for difference in retinal thickness. A comparison
between ranks was evaluated after applying a Bonferroni correction to account for
the number of possible combination of pairs. More animals need to be added to the
survival groups to be able to run a statistical analysis.

3.5 Results
3.5.1 Non-Survival
Thirty-eight rabbits were imaged. The electrode was placed on the inferior
temporal region, close to the visual streak.

3.5.1.1 Electrode-Retina Distance
OCT images were taken as the electrode was advanced towards the retina.
After insertion into the eye, the electrode was advanced from the periphery to a
109
point where the electrode tip was visible in the fundus view of the OCT system. The
electrode was placed closed to the epi-retinal surface but not touching it. The
micromanipulator was used to advance the electrode until it was near the retina.
During the experiment, the electrode was advanced closer to the retina or moved
further from the retina depending on the physiological parameters. Increase of
heart rate or increased of breathing caused the electrode to move closer to the
retina, which it was fixed by moving the electrode with the micromanipulator.
Changes in physiological parameters were controlled by anesthetic
supplementation.

The stimulating electrode was placed inside the eye 117.41±35.65 !"
(mean, SD) away from the retina, measured as shown in Figure 33.

After performing a non-parametric partial correlation analysis between
distance and change in retinal thickness, it was observed that the distance at which
the electrode is placed does not predict the changes in retinal thickness after
accounting for the effects of charge density and stimulating frequency (r=0.044,
p=0.805) (Figure 36)
110

Figure 36. Correlation analysis between distance and percentage of change in retinal
thickness. Distance at which the electrode is placed does not predict the changes in
retinal thickness.

Ten animals were used to study if the distance between the electrode and
the retina affects the time when the change in retinal thickness is observed.
Distance affects the time in which the change in retinal thickness is observed as we
can see in Figure 37. If the distance between the electrode and the retina is less than
100 !" the changes in retinal thickness start being observed within five minutes of
stimulation. In the other hand, if the distance between the electrode and the retina
is more than 100 !" the changes in retinal thickness start being observed between
twelve and fifteen minutes after stimulation was started. Both groups (!"#$%&'(<
100 !" !"# !"#$%&'( > 100 !") contain different combinations of stimulation
111
parameters and it can be observed in Figure 37 that they do not play a role in the
time when the changes in retinal thickness start.

Figure 37. Change of retinal thickness during stimulation overtime. Retinal thickness
changes for different charge density groups and different frequencies are plotted in
this graph. We can observe how charge density and frequency are not the affecting
factor for when changes in retinal thickness start to be observed. Only distance is the
important factor.

3.5.1.2 Stimulation-Retinal Thickness
A fundus image and OCT images were taken before stimulation as control.
During the stimulation period, images were acquired every 3 minutes for a period
of 30 minutes and every 5 minutes until the stimulation was over. After stimulation
112
was over, the electrode was moved away from the retina and images were acquired
every five minutes for a period of fifteen minutes.

The effects of the different charge densities applied to the retina are shown
on (Figure 38 - Figure 45). Each figure is a collection of the images acquired from one
animal per stimulating group. One fundus image and three OCT images are shown
per each stimulating group. The OCT images correspond to one before stimulating,
one 15 minutes into stimulation and one fifteen minutes after we finished
stimulating. Some figures show an FA image.

Figure 38. Results group 1. Stimulation parameters are: charge density
0.92 !" !"
!
, frequency 300 hz, duration of stimulation 30 minutes, biphasic,
cathodic first 1 ms pulses. Top left image. Fundus image of the rabbit’s eye during
electrode insertion. Top middle image. Fundus image showing electrode placement
after it was advanced close to the retina. Top right image. FA taken after stimulation
where no vessel leakage is observed. Bottom left image. OCT image taken before
stimulation and electrode was placed. Bottom middle image. OCT image taken 15
113
minutes into stimulation. Yellow arrow shows the electrode tip. Bottom right image.
OCT image taken 15 minutes after stimulation was over. No changes in retinal
thickness were observed in this group.

Figure 39. Results group 2. Stimulation parameters are: charge density
1.22 !" !"
!
, frequency 100 hz, duration of stimulation 100 minutes, biphasic,
cathodic first 1 ms pulses. Top left image. Fundus image of the rabbit’s eye during
electrode insertion. Bottom left image. OCT image taken before stimulation and
electrode was placed. Bottom middle image. OCT image taken 15 minutes into
stimulation. Yellow arrow shows the electrode tip. Bottom right image. OCT image
taken 15 minutes after stimulation was over. No changes in retinal thickness were
observed in this group.
114

Figure 40. Results group 3. Stimulation parameters are: charge density
1.22 !" !"
!
, frequency 200 hz, duration of stimulation 45 minutes, biphasic,
cathodic first 1 ms pulses. Top left image. Fundus image of the rabbit’s eye during
electrode insertion. Top middle image. Fundus image showing electrode placement
after it was advanced close to the retina. Top right image. FA taken after stimulation
where no vessel leakage is observed. Bottom left image. OCT image taken before
stimulation and electrode was placed. Bottom middle image. OCT image taken 15
minutes into stimulation. Yellow arrow shows the electrode tip. Bottom right image.
OCT image taken 15 minutes after stimulation was over. Changes in retinal thickness
are shown with blue arrows.

115
Figure 41. Results group 4. Stimulation parameters are: charge density
1.22 !" !"
!
, frequency 300 hz, duration of stimulation 30 minutes, biphasic,
cathodic first 1 ms pulses. Top left image. Fundus image of the rabbit’s eye during
electrode insertion. Top middle image. Fundus image showing electrode placement
after it was advanced close to the retina. Top right image. FA taken after stimulation
where no vessel leakage is observed. Bottom left image. OCT image taken before
stimulation and electrode was placed. Bottom middle image. OCT image taken 15
minutes into stimulation. Yellow arrow shows the electrode tip. Bottom right image.
OCT image taken 15 minutes after stimulation was over. Changes in retinal thickness
are shown with blue arrows.

Figure 42. Results group 5. Stimulation parameters are: charge density
1.63 !" !"
!
, frequency 20 hz, duration of stimulation 100 minutes, biphasic,
cathodic first 1 ms pulses. Top left image. Fundus image showing electrode placement
after it was advanced close to the retina. Bottom left image. OCT image taken before
stimulation and electrode was placed. Bottom middle image. OCT image taken 15
minutes into stimulation. Yellow arrow shows the electrode tip. Bottom right image.
OCT image taken 15 minutes after stimulation was over. No changes in retinal
thickness were observed in this group.
116

Figure 43. Results group 6. Stimulation parameters are: charge density
1.63 !" !"
!
, frequency 100 hz, duration of stimulation 100 minutes, biphasic,
cathodic first 1 ms pulses. Top left image. Fundus image of the rabbit’s eye during
electrode insertion. Top middle image. Fundus image showing electrode placement
after it was advanced close to the retina. Bottom left image. OCT image taken before
stimulation and electrode was placed. Bottom middle image. OCT image taken 15
minutes into stimulation. Yellow arrow shows the electrode tip. Bottom right image.
OCT image taken 15 minutes after stimulation was over. Changes in retinal thickness
are shown with blue arrows.

117

Figure 44. Results group 7. Stimulation parameters are: charge density
1.63 !" !"
!
, frequency 200 hz, duration of stimulation 45 minutes, biphasic,
cathodic first 1 ms pulses. Top left image. Fundus image showing electrode placement
after it was advanced close to the retina. Bottom left image. OCT image taken before
stimulation and electrode was placed. Bottom middle image. OCT image taken 15
minutes into stimulation. Yellow arrow shows the electrode tip. Bottom right image.
OCT image taken 15 minutes after stimulation was over. Changes in retinal thickness
are shown with blue arrows.

Figure 45. Results group 8. Stimulation parameters are: charge density
1.63 !" !"
!
, frequency 300 hz, duration of stimulation 30 minutes, biphasic,
118
cathodic first 1 ms pulses. Top left image. Fundus image of the rabbit’s eye during
electrode insertion. Top middle image. Fundus image showing electrode placement
after it was advanced close to the retina. Top right image. FA taken after stimulation
where no vessel leakage is observed. Bottom left image. OCT image taken before
stimulation and electrode was placed. Bottom middle image. OCT image taken 15
minutes into stimulation. Yellow arrow shows the electrode tip. Bottom right image.
OCT image taken 15 minutes after stimulation was over. Changes in retinal thickness
are shown with blue arrows.

Figure 46 summarizes the results of testing the various frequencies and
charge densities. When 1 ms pulses were applied, groups under the next conditions
showed changes in retinal thickness: Charge density: 1.63 !".!"
!
with frequency:
333 !", Charge density: 1.63 !".!"
!
with frequency: 200 !", Charge density:
1.63 !".!"
!
with frequency: 100 !" , Charge density: 1.22 !".!"
!
with
frequency: 333 !" and Charge density: 1.63 !".!"
!
with frequency: 200 !". The
remaining groups stimulated with 1 ms pulses showed no difference in retinal
thickness before and after stimulation: Charge density: 1.63 !".!"
!
with
frequency: 20 !", Charge density: 1.22 !".!"
!
with frequency: 100 !" and
Charge density: 0.92 !".!"
!
with frequency: 333 !". Only one group was tested
using 25 ms pulses at a charge density of 2.26 !".!"
!
and a frequency of 16 !",
and no changes in retinal thickness were observed.

119
It is important to note that all the stimulation parameters tested during this
study are much higher than the parameters used currently in patients implanted
with the Argus II system and this is represented in the figure by the red circle.

Figure 46. Summary of retinal damage for all different groups evaluated. Retinal
damage was observed at 1.63 mC.cm-2 (333, 200, 100 Hz) and 1. 22 mC.cm-2 (333,
200 Hz). Red dot shows the parameters used for Argus II.

After performing a non-parametric partial correlation analysis between
frequency and charge density versus percent change in retinal thickness, it was
observed that both variables are predictors of the percent change in retinal
thickness (r=0.348, p=0.046 and r=0.542, p=0.01 respectively) (Figure 47)

120

Figure 47. Scatter plot showing the correlation analysis between frequency (left
image) and charge density (right image) vs. percentage of change in retinal thickness.
It is observed that both variables are predictors of the percentage change of retinal
thickness.

From the eight groups tested at 1 ms pulses, five of them showed a change in
retinal thickness. At a charge density of 1.63 !".!"
!!
and a frequency of 300 Hz,
retinal thickness increased (median and interquartile range) from
159 !" (144.5 !" !" 162.25 !") to 189 !" (171.75 !" !" 190.75 !"), a 15.96%
(14.23% to 16.69%) increment compared to baseline. At a charge density of
1.63 !".!"
!!
and a frequency of 200 Hz, retinal thickness increased (median and
interquartile range) from 160 !! ( 155 !" !" 165 !" ) to
210 !" (177 !" !" 243 !"), a 22.26% (12.43% to 32.10%) increment compared
to baseline. At a charge density of 1.63 !".!"
!!
and a frequency of 100 Hz, retinal
thickness increased (median and interquartile range) from
157 !" (137.5 !" !" 157 !" ) to 187 !" (166.5 !" !" 215 !" ), a 22.99%
121
(13.16% to 26.94%) increment compared to baseline. At a charge density of
1.22 !".!"
!!
and a frequency of 300 Hz, retinal thickness increased (median and
interquartile range) from 160 !" ( 153 !" !" 163 !" ) to
189 !" (180 !" !" 204 !"), a 16.67% (12.85% to 22.11%) increment compared
to baseline. At a charge density of 1.22 !".!"
!!
and a frequency of 200 Hz, retinal
thickness increased (median and interquartile range) from
154 !" (142 !" !" 166 !") to 219 !" (165 !" !" 237 !"), a 29.68% (13.94%
to 29.96%) increment compared to baseline. The other three groups showed no
change in retinal thickness (Figure 48).

Performing a non-parametric signed-rank test, the median difference
between the retinal thickness before and after simulation is 23. The p-value from
the Signed Rank test is <0.0001. Thus, we have statistical significance that the
median difference between the retinal thickness before and after simulation is
different from zero.

Since there are 8 different stimulating groups (varying charge density and
frequency), a new variable called treatment was created that includes the different
combinations of charge density and frequency. To assess if the treatment groups
have an effect on the difference in retinal thickness, a non-parametric ANOVA-
Kruskal Wallis test was performed. There is a statistical significance indicating that
the median change of retinal thickness of the 8 groups of charge density and
frequency is different (p=0.0005). A non-parametric post-hoc test using the
122
Wilcoxon test was performed. Ranks were assigned to the data for difference in
retinal thickness. A comparison between ranks was evaluated after applying a
Bonferroni correction to account for the number of possible combination of pairs. A
total of six different combinations were found significantly different: at a frequency
of 300 Hz, 0.92 !".!"
!!
vs 1.22 !".!"
!!
: p=0.00004, at a frequency of 300 Hz,
0.92 !".!"
!!
vs 1.63 !".!"
!!
: p=0.0011, at a frequency of 100 Hz, 1.22
!".!"
!!
vs 1.63 !".!"
!!
: p=0.00057, at a charge density 1.22 !".!"
!!
, 100 Hz
vs 200 Hz: p=0.00041, at a charge density 1.22 !".!"
!!
, 100 Hz vs 300 Hz:
p=0.0010, at a charge density 1.63 !".!"
!!
, 20 Hz vs 100 Hz: p=0.0057. Two
combinations were found to be marginally significantly different: at a charge
density 1.63 !".!"
!!
, 20 Hz vs 200 Hz: p=0.054, at a charge density 1.63
!".!"
!!
, 20 Hz vs 300 Hz: p=0.066. The rest of the combinations didn’t show
differences.

123

Figure 48. Box plot showing the summary results of the eight groups tested. On each
box, the central mark indicates the median, and the bottom and top edges of the box
indicate the 25th and 75th percentiles, respectively. The whiskers extend to the most
extreme data points not considered outliers, and the outliers are plotted individually
using the '+' symbol.

We attempted to fit a logistic regression model after transforming the
percent change variable from continues to logical by considering percentage
changes of retinal thickness greater than five percent as one, and lower than five
percent as zero. Given the low independent, we found that the estimation of the
regression coefficients were not reliable.

124

3.5.2 Survival
Five more rabbits followed the same stimulation protocols described above
but they were kept for a period of two weeks after stimulation. All five rabbits were
stimulated with parameters that caused changes in retinal thickness. OCT images
were acquired every 3 days during the two week period and retinal thickness was
measured like shown in Figure 34. Figure 49 shows images acquired from one
animal. 6 oct images are presented. One before the stimulation, one fifteen minutes
during stimulation, one fifteen minutes after stimulation, one three days after
stimulation, one nine days after stimulation and one twelve days after stimulation.

Figure 49. OCT images of the retina during electrical stimulation and follow-ups.
Stimulation parameters of this animal are: charge density 0.92 mC/cm
2
, frequency
125
333 Hz, duration of stimulation 30 minutes, biphasic cathodic first 1 ms pulses. Top
left image was taken before stimulation was started. Middle left image was taken 15
minutes into stimulation. Bottom left was taken 15 minutes after stimulation was
stopped. Top right image is from the first follow up 3 days after stimulation. Middle
right was taken 9 days after stimulation. Bottom right was taken twelve days after
stimulation. Yellow arrow indicates the tip of the stimulating electrode. Blue Arrows
indicate the area of retinal thickness increment after stimulation.

In all five animals we observed the change in retinal thickness remained
throughout the twelve days. We observed in all animals that three days after
stimulation the change in retinal thickness was increased compared to the last
measurement taken after stimulation. During the second set of OCTs post
stimulation done six days after stimulation the retinal thickness decreases
compared to the third day measurement and after that, we can observe in the
measurements acquired nine and twelve days after stimulation, the changes in
retinal thickness stabilizes. The change in retinal thickness twelve days after
stimulation is between 15 percent and 40 percent for all animals comparable to the
initial retinal thickness (Figure 50).
126

Figure 50. Percentage of change of retinal thickness during stimulation overtime.
Retinal thickness changes for different charge density groups and different
frequencies are plotted in this graph. Changes in retinal thickness remained
throughout the twelve days for all groups.

3.6 Conclusion
Electrical stimulation at very high charge densities causes swelling of the
retina within minutes of stimulation. OCT images showed swelling of within several
retinal layers. The stimulus levels at which swelling was noted is significantly
higher than parameters used generally for stimulation in humans in terms of rate
and charge density, but in limited cases high currents [134] or high rates [135]
have been used in humans, on a short term basis in the context of controlled
psychophysical testing. Other studies have linked stimulation rate and/or duty
127
cycle to neural damage [136]. Rate may be an important factor for retinal
stimulation safety as well.

It is not clear why electrical stimulation causes damage to the tissue. Some
hypothesis states that intrinsic biological processes cause tissue damage when the
tissue is over stimulated. Over stimulation of the tissue can cause neuron firing for
long periods of time, changing ion concentration intracellular and extracellular,
depletion of oxygen and glucose, excessive release of glutamate, etc creating a toxic
environment for the neuron. Retinal damage is also attributed to toxic
electrochemical reactions caused between electrode – tissue interface, when
stimulation is at a rate greater than what the physiological system can
tolerate[126]. While we did not notice bubble formation at the electrode, which
would indicate hydrolysis, we cannot rule out other chemical reactions that may
have altered PH or produced harmful reactants.

We developed a new method that allows us to study the effects of electrical
stimulation on the retina in an in vivo model during the stimulation period. We
used an in vivo model because it allows us to study the retina response in its
natural conditions. The technique will allow the study of long-term effects of retinal
stimulation. This is important because studies will be needed to see if this swelling
is temporary or permanent helping us understand its causes and potential effects
on retinal implant patients.
128
From this study we can conclude that frequency and charge density play an
important role in the change of retinal thickness during stimulation. In addition,
distance between electrode and retina affects the time when the retinal increment
is observed. Lastly, retinal changes in thickness are still observed two weeks after
stimulation.

Finding that charge density, frequency and distance play a very important
role in the effects of electrical stimulation in retinal tissue, we tried to create a
mathematical model that will help us predict the safe and unsafe levels but during
this process we found that more measurements need to be added to be able to
create a reliable model.

Based on our experimental results and previous research such as the one
carried out by Shannon [130], we believe that it would be very important for the
field to create a mathematical model that will help find the safe and unsafe
electrical stimulation parameters for neural tissue. In addition to testing more
combinations of frequency, charge density and distance, future studies should
explore the effects that variations in electrode size and duty cycle have on tissue
damage. We believe that our results bring us one step closer to create a strong and
reliable model that will enable the prediction of damage in retinal and neural
tissue.

129
CHAPTER 4
Test Stimulus Protocols in Retinal-Prosthesis Patients

3.1 Background
3.1.1 Retina response to electrical stimulation

The retina is made out of multiple and different type of cells as explained in
chapter 1. Between the photoreceptors and the ganglion cells, there are three
different types of neural cells: horizontal cell, amacrine cells and bipolar cells,
which combine signals from different photoreceptors creating precise spatial and
temporal patterns, which are crucial for the responses evoked by the ganglion cells.

On-center ganglion cells is excited when light stimulates the center of its
receptive field and inhibited when light stimulates its surround. An off-center
ganglion cells exhibits an opposite behavior, by being excited when light stimulates
its surround and inhibited when light stimulates its center. They are present
roughly in equal numbers, and every photoreceptor send output to both types,
causing ganglion cells to produce two parallel pathways for the processing of visual
information. Bipolar cells are also classified as on-center or off-center. When cones
in the center of the receptive field are active, on-center bipolar cells depolarize,
130
while off-center bipolar cells hyperpolarize. When cones in the surround are active,
the response of the bipolar cells is opposite that evoked by illumination of the
center [2].

When an stimulus pulse is sent to the retina while recording from the
ganglion cells, it has been found that ganglion cells produces two different types of
latencies (Latency refers as the time between the stimulus is sent and the first cell
spoke is observed). Short latency classifies as less than 2-5 ms and long latency
greater than 2-8 ms; if a synaptic blocker is used to eliminate the effects of
presynaptic cells, long latencies disappear, demonstrating that long latencies are
attributed to presynaptic cells response, and long latencies to direct activation of
ganglion cells [137] . In contrast with this study, Sekirnjak et al. observed a paired
response (both short and log latency) to stimulation with direct activation of
retinal ganglion cells (RGCs) [84]. This is important because being able to target
neural cells based on their specificity will allow the creation of a retinal prosthesis
with better control over the subject’s perception [138] [139].

The ganglion cell response to electrical stimulation can be classified as
direct activation and indirect activation by stimulating second order neurons.
Ideally retinal stimulation will be able to produce ganglion cell spiking patterns
that replicate the temporal and spatial specificity of normal vision.

131
Direct ganglion cell activation generates short latency responses, (~ 5 ms)
and it usually elicits a single spike per stimulus when stimulated epiretinally or
subretinally. It seems like short pulse width stimulation (<0.4 ms) is more prone to
generate direct ganglion cell activation [140] but also has the potential to
antidromically activate peripheral RGCs by stimulating axons pasing beneath the
electrode [141].

Indirect ganglion cell activation occurs through the stimulation of secondary
neurons that subsequently release synaptic neurotransmitters onto RGC dendrites
generating a more natural response of the RGC cells. Indirect stimulation typically
elicits a burst of spikes and their latencies are longer than those for direct
activation. The definition of long latency-spikes is variable but generally means that
the response peaks tens of milliseconds after stimulus onset [142]. Some studies
have shown that long pulse width stimulation (>25 ms) elicits indirect responses
without activating underlying axons, causing the activation to stay close to the
electrode [143].

3.2 Previous Research
Multiple studies in vitro and in patients have been reported to better
understand the spatial patterns of RGC activation.

132
Weitz et al. developed an in vitro retinal preparation to investigate stimulus
paradigms. They designed an adeno-associated viral (AAV) vector that transduces
more than 80% of RGCs in the rat retina with the genetically encoded calcium
indicator GCaMP5G [143]. Calcium imaging was used to record from RGC
populations in isolated retinas while stimulating with transparent multielectrode
arrays (MEAs) allowing real-time mapping of RGC activity both in a large area and
at single-cell resolution [141].

Different pulse widths were studied to see how they affected the spatial
extend of RGC activation. All pulses 8 ms and shorter activated axons, causing
streak responses extending from the electrode to the edge of the retina. Sixteen-
millisecond pulses also stimulated axons but to a much lesser extent (fewer than
5% of the cells ≥100 mm to the right of the electrode perimeter responded to
stimulation). Pulses 25 ms and longer produced no evidence of axonal stimulation,
instead resulting in focal activation. Threshold charge density became higher as
pulse width increased, indicating that longer pulses were less efficient at evoking
responses [82] (Figure 51) .

Because high-resolution implants will need smaller electrodes, different
diameter electrodes were tested at the pulse widths mentioned above. They found
that threshold charge density increased as electrodes became smaller, whereas
response shape remained the same. The size of the response area decreased with
electrode size but only to a certain extent. With 25-ms pulses, for example, 75-um
133
electrodes evoked responses from a smaller area than 200-um electrodes, but no
further reduction in response area was achieved with 30-um electrodes (Figure 51).

The conclusion of this study was that clinically approved retinal implants
use stimulus pulses in the order of 1 ms, that based on their results doesn’t
discriminate the activation of RGC axons, while long pulses target bipolar cells
rather than photoreceptors.

134

Figure 51. Spatial threshold maps for three pulse durations that cover the gamut of
response types: direct RGC stimulation (0.06 ms, left column), combined ganglion and
bipolar cell stimulation (1 ms, middle col- umn), and bipolar cell stimulation (25 ms,
right column) in RD and wild-type (WT) rats. In all cases, 0.06-ms pulses provided
good selectivity for local somata over passing axons, 1-ms pulses provided poor
selectivity and 25-ms pulses produced focal responses. The bottom row shows
background-subtracted GCaMP5G responses to suprathreshold stimuli for 30 um-
diameter electrodes (black circles). Image taken from [141].

The conclusion of Weitz’s study brings us to a study performed by Nanduri
et al. where they manipulated pulse train frequency and amplitude in a patient
implanted with a 16- channel epiretinal prosthesis (Argus I, Second Sight Medical
Products). For the modulated- amplitude experiment, current amplitude was
135
modulated between 1.2X and 6X threshold while holding the frequency constant at
20 Hz. In the modulated- frequency experiment, the frequency of the pulse train
was varied between 13 Hz and 120 Hz, while current amplitude was held constant
at 1.25X threshold. Phosphene shape was measured by asking the subject to outline
the shape on a grid screen (containing 6-in. horizontal and vertical gridlines) with a
center location aligned horizontally and vertically with the subject’s head. Drawing
was performed with a pen that had a cap of a different color from the pen. A head-
mounted camera (CMOS S588-3T; Misumi, Tokyo, Japan), located on the subject’s
glasses, was used to record the trials to a digital video recorder (DVR). Video files
were analyzed off-line to extract shape data using custom-built tracking software.
Phosphene brightness was measured with a brightness rating procedure that
consisted on the subject comparing the brightness of the phosphene to a reference
stimulus. Subject was instructed to rate the apparent brightness independently
from the apparent size of the phosphene [80].

This study shows that manipulating current amplitude and frequency have
diferent effects on phosphene size and brightness. Changing current pulse
amplitude affects both phosphene area and, to a lesser degree, phosphene
brightness. In contrast, changing current pulse frequency results in relatively little
change in area, but a much larger change in brightness (Figure 52).

136

Figure 52. The effects of amplitude and frequency on apparent brightness and size.
(A, C) Brightness and apparent size as a function of normalized (relative to threshold)
amplitude for nine electrodes. (B, D) Brightness and apparent size as a function
frequency for the same nine electrodes. Each electrode’s data are fit with the best-fit
linear regression. Figure taken from [80].

Other experiment performed in 4 subjects, 1 implanted with an Argus I
system and 3 implanted with the Argus II system, looks to explore the shapes of
single electrode phosphenes by performing direct single electrode stimulation
experiments and asking patients to draw in a board or a touch screen the shape
they perceived [81]. Results show that for each individual electrode patients
reports different shapes like straight lines of varying thickness, ovals, round spots
or wedges (Figure 53). There was not a relationship between phosphene size and
stimulus location relative to the fovea. Likely, a perceived phosphene is result of
response from a population of neural cells (and passing axon fibers) located
137
directly below the stimulating electrode, which verifies the in vitro results
encountered by Weitz [143].

Figure 53. Drawings from single electrodes overlayed on electrode array for each one
of the patients tested. Red circles indicates stimulating electrode, blue square marks
the approximate location of the fovea and black marks represents the subjects
drawing to that specific electrode.

3.3 Specific Aim
Subjects’ response to the epiretinal implant has been extensively studied.
Experiments where the variation of the pulse width stimulation has not been
studied yet because the current video processing unit (VPU) of the Argus II system
doesn’t allow these changes. A new experimental VPU has been developed, that will
138
allow us to perform different protocols with patients, but before we start modifying
parameters, a baseline data needs to be acquired to compare results.

3.4 Methods
3.4.1 System

All patients have been implanted with the Argus II system from The Second
Sight Medical Products Inc. More details about the system are in chapter 1, section
1.7.1.1.

The implanted pulse generator (IPG) of the system produces the
commanded stimulus pulses and transmits it to the electrode array. Stimulation
can be presented in two different forms:
1) Camera mode – video captured by the camera is continuously sampled by the
VPU to match the stimulation current amplitude in each electrode to the brightness
of the image captured.
2) Direct stimulation mode - the stimulation signal sent to each electrode is
independently controlled by the VPU. The VPU is connected to a computer that
allows the researcher/physician select the desired stimulus.
All experiments in this thesis were performed in direct stimulation mode.

139
3.4.2 Subjects

We performed our experiments in 6 different subjects implanted with the
Argus II system device (Second Sight Medical Products Inc., Sylmar, CA). All tests
were performed after obtaining informed consent under a protocol approved by
the Institutional Review Board (IRB) at each subject location and under the
principles of the Declaration of Helsinki. The University IRB approved subject 5
while the University of Michigan IRB approved Subject 1-4.

3.4.2.1 Subject 1
Subject 1 is male, age 68, and was implanted in August 20
th
, 2015 with the
Argus II device in his left eye at Kellogg Eye Center in The University of Michigan.
The subject had no light perception in his left eye and light perception without
projection in his right eye pre-operatively, and has been blind since August 1982,
caused by retinitis pigmentosa. Besides the system implantation, the subject had
cataract surgery in both eyes. Subject reports using the system five times a week
(Monday – Friday). Test was carried out in 2017

3.4.2.2 Subject 2
Subject 2 is female, age 68, and was implanted in February 20
th
, 2014 with
the Argus II device in her left eye at Kellogg Eye Center in The University of
Michigan. The subject had light perception in both eyes pre-operatively but during
140
the test reports no light perception in left eye and light perception in the periphery
of the right eye. Subject has been totally blind since 2001 caused by retinitis
pigmentosa. Besides the system implantation, the subject had vitrectomy and
wound revision surgery in 2014, and cataract surgery of the left eye. Subject
reports using the system three times a week. Test was carried out in 2017

3.4.2.3 Subject 3

Subject 3 is female, age 68, and was implanted in May 18
th
, 2015 with the
Argus II device in her right eye at Kellogg Eye Center in The University of Michigan.
The subject had light perception in both eyes pre-operatively. Subject was
diagnosed with retinitis pigmentosa at 35 years of age and went blind at 47.
Besides the system implantation, the subject had cataract surgery in both eyes.
Subject had a posterior vitreous detachment and has also being diagnosed with
Charles Bonnet Syndrome. Subject reports using the system three times a week.
Subject reports using the system five times a week. Test was carried out in 2017

3.4.2.4 Subject 4
Subject 4 is male, age 69, and was implanted in July 16
th
, 2015 with the
Argus II device in his left eye at Kellogg Eye Center in The University of Michigan.
The subject had light perception in both eyes pre-operatively. Subject has been
totally blind since 1996 caused by retinitis pigmentosa. Besides the system
implantation, the subject had cataract surgery in both eyes and has been diagnosed
141
with filamentary keratitis. Subject reports using the system once a week. Test was
carried out in 2017

3.4.2.5 Subject 5
Subject 5 is male, age 67, and was implanted in 2004 with the Argus I device
in his right eye and in 2015 with the Argus II device in his left eye at The University
of Southern California. The subject had no light perception in both eyes pre-
operatively. Subject was diagnosed with retinitis pigmentosa in 1985, and was
blind for 10.5 years before the implantation of the first system. Besides the system
implantation, the subject had surgery to fix a retinal detachment in 2009. Subject
reports using the system five times a week. Test was carried out in 2016

3.4.3 Psychophysical Tests
Patients were asked to perform a drawing task either by presenting them a
shape or when their retina was electrically stimulated. The same test was
presented in multiple trials in a random way to confirm reproducibility of results.

3.4.3.1 Tactile – Drawing Task
Stimulation – Drawing task relies on the subject’s ability to draw percepts
accurately and consistently across trials. As seen in the subject’s description, all of
them have been blind for many years. The lack of tactile- visual feedback for many
142
years can cause variations in their drawings and this is why a control experiment is
performed with tactile targets to establish a baseline drawing error for each
subject.

We used 12 different shapes made out of cardboard (Figure 54). Subjects
were asked to feel the shape, and then draw them on a touch screen and describe
verbally what shape they thought the target was. Touch screen data was instantly
recorded by custom software provided by Second Sight as a text file. The text file
was analyzed offline with a program written in Matlab.

Figure 54. Shapes presented to patients during tactile control experiment.

143
3.4.3.2 Stimulation – Drawing Task
Subject was placed in a chair at a comfortable distance from a touchscreen
monitor. The distance from the subject’s eyes to the screen was approximately 40
inches (Figure 55). One stimulus was presented to the subject, and the subject was
asked to trace the shape on the monitor and to describe verbally the shape, size,
color, and brightness of the percept. Then the experiment was advanced to the next
trial. Touch screen data was instantly recorded by custom software provided by
Second Sight as a text file. The text file was analyzed offline with a program written
in Matlab.

Figure 55. Left Image. Example of an Argus II subject drawing task. Right Image.
Subjects drawing recorded by a touchscreen monitor.

A computer controlled direct stimulation testing with a graphic user
interface (GUI) created in Matlab. The GUI allows us to upload the subjects latest
VCF which will automatically enter to the system the latest threshold and settings
of each individual patient. Electrodes with unsafe thresholds or any other issue will
be blocked from the program and are not used during our experiments. A grid
144
representing the 60 electrodes allows us to select the individual electrodes or the
pattern we want to present to the subject. All the stimulation patterns for each
patient are programmed previously and uploaded to the system. The system selects
randomly the pattern presented to the subject. The parameter video level controls
the brightness of the stimulus presented to the subject. Experiments are started
with a video level of 10 and increased to a maximum of 31 depending of what the
patient reports as comfortable and visible. Duration of the pulse is set at 250 ms,
but it is changes to 1000 ms before experiments are started. The rest of the
parameters are left as default (Figure 56)

Figure 56. Pattern stimulation GUI where stimulation is controlled and parameters
can be varied.

145
3.4.3.3 Retinal stimulation parameters
For baseline experiments, tasks were performed under the Argus II normal
operation parameters. Stimuli were charged-balanced, 0.45 ms/phase cathodic-
first biphasic pulse train with a total stimulus duration of 1000 ms and a frequency
of 6 Hz. Pulses were charge balanced. Two different types of experiments were
tested: 1) Single electrode stimulation, 2) Simple pattern/Letter stimulation. For
both groups, each stimulus was presented 3 times in random order.

During single electrode stimulation, each electrode will be stimulated
independently three times and the patient was asked to draw in the touch screen
what they perceived. During pattern stimulation, different letter shapes (L, T, V, H,
C, O, D and P) will be presented in each trial. Each trial will be picked in random
order.

3.4.4 Analysis
The location of the fovea was estimated examining the subject’s fundus
photograph. The electrode array and the optic nerve are visible in the fundus image
of each subject, which allows us to use them as markers. The fovea is approximated
to be temporal and inferior to the optic disc. Thus the fovea is approximated at an
angular distance of 15.5±1.1° (temporal to the center of the optic disc), and at a
vertical angular distance of -1.5±0.9° (inferior to the center of the optic disc) as
shown in Figure 57.
146

Figure 57. Left Image. Fundus photograph. Right Image. Big red circle is marking the
optic nerve, and small circle is marking the foveal region.

All text files are analyzed in MATLAB. The fovea region is marked in each of
the files created in matlab where we are able to see the stimulus presented to the
subject and their response to each time that same stimulus was presented.

3.5 Results
3.5.1 Subject 1

During the experiment patient reported having a lot of spontaneous retinal
responses when he had the system on causing him a lot of background noise. This
background noise made it difficult for him sometimes to distinguish between
spontaneous responses and actual stimulations. During the experiment subject
asked multiple times to make the phosphenes brighter so he could see them better.
147
Video level was set at 31, which it was the maximum brightness we could provide
to him with the system.

Figure 58 is a fundus image of the subject’s implanted eye and it was
provided by the clinic. We found the fovea and the angle of the electrode array
using this image and we used this information for the analysis of the results.

Figure 58. Left Image is a fundus image of the retina of subject 1 implanted eye. Optic
nerve and electrode array are used as landmarks to find the angle and location of the
fovea. Right image shows in a big red circle the optic nerve and a small red circle the
fovea region. The electrode array is overlapped with the array generated by matlab
for comparison.

Percepts varied for each electrode, but were relatively consistent from trial
to trial. Subject reported that percepts were bright flashes of light that after
multiple trials started to appear more dull. He reported to see white flashes of light.
Figure 59 shows the results of twelve electrodes during single electrical
stimulation experiments that were picked randomly to represent different sections
148
of the array. The left black square shows the electrode array with a white circle that
represents the electrode stimulated during that trial. The array is positioned in the
angle found with the fundus image and the red circle represents the fovea location.
The right image right next to it represents the patients’ response to the stimulation
of that specific electrode. Subject reported seeing circles, ovals, lines and flashes of
light with no specific shape when electrodes were stimulated independently.

Figure 59. Drawing from 12 single electrodes are shown in this image. Images with
black background show the stimulus presented to the patient, this image is tilted to
represent the tilt of the array in the retina. Images in white background show the
patient’s drawing when that stimulus was presented. The red circle represents the
location of the fovea

149
During pattern stimulation subject is asked to describe what type of pattern
he thinks it is being presented to him. Subject is not able to recognize any of the
patterns presented during any of the trials. Subject reports that these patterns are
bigger and brighter than the ones previously presented (single electrode
stimulation trials), and describes all of them as big flashes of light we no specific
shape. Figure 60 is showing the results for the pattern stimulation experiments
and it is in the same format described above. Simple patterns like a square, a line
and a diagonal line were presented to the subject and the letters L, C, D, T, H, O and
A.

Figure 60. Drawing from 11 different patterns presented to subject 1. Images with
black background show the pattern presented to the patient, this image is tilted to
150
represent the tilt of the array in the retina. Images in white background show the
patient’s drawing when that stimulus was presented. The red circle represents the
location of the fovea

3.5.2 Subject 2

During the experiment patient reported her system was not working
correctly and sometimes she couldn’t see anything with the system ON. In the
beginning of single electrode stimulation trials, patient reported multiple times she
couldn’t see anything but it was not related with the system not working correctly.
Video level was set at 31, which it was the maximum brightness we could provide
to her with the system.

Figure 61 is a fundus image of the subject’s implanted eye and it was
provided by the clinic. We found the fovea and the angle of the electrode array
using this image and we used this information for the analysis of the results.

Figure 61. Left Image is a fundus image of the retina of subject 2 implanted eye. Optic
nerve and electrode array are used as landmarks to find the angle and location of the
151
fovea. Right image shows in a big red circle the optic nerve and a small red circle the
fovea region. The electrode array is overlapped with the array generated by matlab
for comparison.

During tactile experiments, subject 2 was able to draw all eleven shapes
presented to her in each trial. Being able to perform this experiment accurately,
suggests that during the stimulation experiment, drawing the correct/incorrect
shape has a smaller chance of being a drawing error. Figure 62 shows the results of
ten shapes presented to the subject and her response. The left green image shows
the shape presented to the subject. The right image right next to it represents the
patients’ response to the shape presented.

Figure 62. Green images are the shapes presented to the subject. Subject was asked
to feel them with both hands and report in the touchscreen monitor the shape of the
object presented. The white image is the patient’s response to the shape. She was able
to draw all shapes consistently and accurately.

152
For 21 out of the 60 electrodes, subject 2 reported no percepts during single
electrode stimulation for all trials. For the rest of the electrodes the percepts
shaped varied for each electrode, but were relatively consistent from trial to trial.
Subject reported that percepts were sometimes bright flashes and sometimes very
small dull sports of light very unnoticible. Figure 63 shows with red exes the 21
electrodes that did not produced a phosphene and the results of extra nine
electrodes during single electrical stimulation experiments that were picked
randomly to represent different sections of the array. All images use the same
format described in patient 1. Subject reported seeing circles, ovals, lines, squares,
semicircles and small dots. It is important to note that this subject was very careful
and paid special attention to draw phosphenes exactly how she perceived them.

153
Figure 63. Top left array shows 21 x’s representing the 21 electrodes in which this
subject reported not seeing phosphenes when stimulated. 9 extra electrodes were
randomly picked to show their results. Images with black background show the
stimulus presented to the patient, this image is tilted to represent the tilt of the array
in the retina. Images in white background show the patient’s drawing when that
stimulus was presented. The red circle represents the location of the fovea

During pattern stimulation the experiment was varied for this subject since
she reported not seeing phosphenes with 21 of the electrodes. Some grouping
experiments were performed, where two or more of the electrodes with not
response during single electrode stimulation were stimulated simultaneously and
as the previous experiment, patient was asked to describe and draw the perception.
When two electrodes were stimulated simultaneously, subject reported seeing a
very small dull light almost impossible to distinguish. The more electrodes were
added to the stimulation, the brighter and bigger the percept became. Big patterns
and small patterns were presented to the subject since a lot of the electrodes had
no response. Subject was not able to recognize any of the patterns presented
during any of the trials. For some shapes like for example the letter L, subject
reports seeing an elongated shape but cannot distinguish the exact shape. Figure
64 shows the results for the pattern stimulation experiments and also the grouping
experiments. The first two rows are examples of the grouping experiments and the
last two are the pattern experiments. Letters like T, H, L, small T, small H, simple
shapes like squares and lines and grouping of two, four and nine electrodes are
reported.
154

Figure 64. Drawing from 12 different patterns presented to subject 2. Images with
black background show the pattern presented to the patient, this image is tilted to
represent the tilt of the array in the retina. Images in white background show the
patient’s drawing when that stimulus was presented. The red circle represents the
location of the fovea. The first two rows are the results of the grouping experiment
and the last two are the results of pattern experiments.

3.5.3 Subject 3
The day of the experiment, the patient commented she was diagnosed with
Charles Bonnet syndrome, which it causes her background light to change colors.
She reports that her background was grey during our experiment so contrast
155
between stimulation and background is not very strong. Video level was set at 31,
which it was the maximum brightness we could provide to her with the system to
provide higher contrast.

Figure 65 is a fundus image of the subject’s implanted eye and it was
provided by the clinic. We found the fovea and the angle of the electrode array
using this image and we used this information for the analysis of the results.

Figure 65. Left Image is a fundus image of the retina of subject 3 implanted eye. This
subject was implanted in the right eye. Optic nerve and electrode array are used as
landmarks to find the angle and location of the fovea. Right image shows in a big red
circle the optic nerve and a small red circle the fovea region. The electrode array is
overlapped with the array generated by matlab for comparison.

During tactile experiments, subject 3 was able to draw all eleven shapes
presented to her in each trial. Being able to perform this experiment accurately,
suggests that during the stimulation experiment, drawing the correct/incorrect
shape has a smaller chance of being a drawing error. Figure 66 shows the results of
ten shapes presented to the subject and her response. The left green image shows
156
the shape presented to the subject. The right image right next to it represents the
patients’ response to the shape presented. Patient was able to report the shape of
the letter H correctly but the subject didn’t lift the finger from the screen while
drawing it and this is what we can observe in the figure.

Figure 66. Green images are the shapes presented to the subject. Subject was asked
to feel them with both hands and report in the touchscreen monitor the shape of the
object presented. The white image is the patient’s response to the shape. She was able
to draw all shapes consistently and accurately.

Nine electrodes were blocked for safety reasons and are reported in the
image below with a yellow x, for other 6 electrodes, subject 3 reported no percepts
during single electrode stimulation for all trials. For the rest of the electrodes the
percepts shaped varied for each electrode, but were relatively consistent from trial
to trial. For some specific electrodes like for example electrode E03, subject
reported seeing colors like orange or pink. Figure 67 shows with red exes the 6
electrodes that did not produced a phosphene and the results of extra nine
157
electrodes during single electrical stimulation experiments that were picked
randomly to represent different sections of the array. All images used the same
format described in patient 1. Subject was very careful with her drawings and
descriptive of phosphenes, reporting mostly very small circles, ovals or squares.

Figure 67. Top left array shows 6 red x’s representing the 6 electrodes in which this
subject reported not seeing phosphenes when stimulated and 9 yellow x’s
representing electrodes blocked by the system for safety reasons. 9 extra electrodes
were randomly picked to show their results. Images with black background show the
stimulus presented to the patient, this image is tilted to represent the tilt of the array
in the retina. Images in white background show the patient’s drawing when that
stimulus was presented. The red circle represents the location of the fovea

158
Subject was able to recognize one shape during pattern stimulation. When
the letter V was presented the subject reported seeing “Maybe the letter L” as
shown in the figure below. For letter P, subject reported “ A line and a circle, maybe
a small I”. For the letter L, a “slanted line” was reported. For the letter T a “very big
and elongated shape” was described for the subject. For the rest of the figures,
subject reports flashes, big circles, and long and elongated lines. Figure 68 shows
the results for the pattern stimulation. Even though subject doesn’t say verbally the
correct letter, her descriptions to the letters for some of them have a lot of
similitudes with the pattern presented.

Figure 68. Drawing from 12 different patterns presented to subject 3. Images with
black background show the pattern presented to the patient, this image is tilted to
159
represent the tilt of the array in the retina. Images in white background show the
patient’s drawing when that stimulus was presented. The red circle represents the
location of the fovea.

3.5.4 Subject 4

Subject has light perception in both eyes. He reported sometimes he has
issues recognizing the stimulation coming from the device for lack of contrast. In
the beginning of single electrode stimulation trials, patient reported multiple times
he couldn’t see the stimulus presented. Video level was set at 31, which was the
maximum brightness we could provide to him with the system. When stimulus was
presented, subject reported stimulations was too fast which it made it difficult for
him to recognize any specific shape. During the second trial duration of the
stimulation was increased to 1500 ms but no changes in the results were observed.

Figure 69 is a fundus image of the subject’s implanted eye and it was
provided by the clinic. We found the fovea and the angle of the electrode array
using this image and we used this information for the analysis of the results.

160
Figure 69. Fundus image of subject 4. Patient was implanted in his left eye. Right
image shows a big red circle showing the optic nerve and a small red circle showing
the fovea region. The electrode array is overlapped with the array generated by
matlab for comparison.

During tactile experiments, subject 4 was able to draw all eleven shapes
presented to him in each trial correctly. Being able to perform this experiment
accurately, suggests that during the stimulation experiment, drawing the
correct/incorrect shape has a smaller chance of being a drawing error. Figure 70
shows the results of ten shapes presented to the subject and his response. The left
green image shows the shape presented to the subject. The right image right next
to it represents the patients’ response to the shape presented. Patient reported
verbally the correct shape for all eleven shapes, but used two hands when asked to
draw in the monitor causing some images not as accurate. After told to only use one
hands, the rest of the results were reported accurately.

161

Figure 70. Shape responses for subject 4. Green images are the shapes presented to
the subject. Subject was asked to feel them with both hands and report in the
touchscreen monitor the shape of the object presented. The white image is the
patient’s response to the shape. He was able to draw all shapes consistently and
accurately.

For 30 out of the 60 electrodes, subject 4 reported no percepts during single
electrode stimulation for all trials. For the rest of the electrodes the percepts
shaped varied between lines and some circles for each electrode, but were
relatively consistent from trial to trial. Subject reported that percepts were too fast
for him to recognize any shapes, he described the percepts as fast flashes of light.
Figure 71 shows with red exes the 30 electrodes that did not produced a
phosphene and the results of extra nine electrodes during single electrical
stimulation experiments that were picked randomly to represent different sections
of the array. All images use the same format described in patient 1. Subject
reported mostly lines for all electrodes.

162

Figure 71. Top left array shows 30 x’s representing the 30 electrodes in which this
subject reported not seeing phosphenes when stimulated. 9 extra electrodes were
randomly picked to show their results. Images with black background show the
stimulus presented to the patient, this image is tilted to represent the tilt of the array
in the retina. Images in white background show the patient’s drawing when that
stimulus was presented. The red circle represents the location of the fovea

As we learned from the single electrode stimulation experiment, only half of
the electrodes from this subject produced phosphenes, and we can see this
reflected in the pattern stimulation experiment, where the subject is not able to
successfully recognize any pattern or shape. For any shape or letter presented, the
subject always reported a long flash of light bigger than the previous percepts
(single electrode stimulation). Figure 72 shows the results for the pattern
163
stimulation experiments. Letters A, C, O, L, H, D, T, simple shapes like a line, big
square, a diagonal and some grouping of electrodes are reported.

Figure 72. Drawing from 12 different patterns presented to subject 4. Images with
black background show the pattern presented to the patient, this image is tilted to
represent the tilt of the array in the retina. Images in white background show the
patient’s drawing when that stimulus was presented. The red circle represents the
location of the fovea.

164
3.5.5 Subject 5

Subject 5 has been implanted with the Argus I system for over 10 years, so
he has a lot of experience running this type of experiments. Video level was set at
31, which it was the maximum brightness we could provide to him with the system.

Figure 73 is a fundus image of the subject’s implanted eye with the Argus II
system, which it is the system we used during this experiment. We found the fovea
and the angle of the electrode array using this image and we used this information
for the analysis of the results.

Figure 73. Fundus image of subject 5. Patient was implanted in his left eye with the
Argus II system. Right image shows a big red circle showing the optic nerve and a
small red circle showing the fovea region. The electrode array is overlapped with the
array generated by matlab for comparison.

165
Percepts varied for each electrode, but were relatively consistent from trial
to trial. Subject reported that percepts were bright flashes of light that after
multiple trials started to appear more dull. He reported to see white flashes of light.
Figure 74 shows a summary of the results found during single electrode and
pattern stimulation. Patterns and electrodes were picked randomly to represent
different sections of the array. The left black square shows the electrode array with
a white circle that represents the electrode stimulated during that trial. The array is
positioned in the angle found with the fundus image and the red circle represents
the fovea location. The right image right next to it represents the patients’ response
to the stimulation of that specific electrode. During single electrode stimulation
subject drew mostly circles and ovals.

During pattern stimulation letters Z, V, L, T, C, H were stimulated and subject
5 was able to identify both verbally and with his drawing the letter L and letter T.
For letter H, he did not report verbally that he was seeing a letter H but he reported
“2 lines with a line cutting them in the middle”. For the letter D, patient reported a
triangle looking shape.

166

Figure 74. Subject 5 responses to single and pattern stimulation. Images with black
background show the stimulus presented to the patient, this image is tilted to
represent the tilt of the array in the retina. Images in white background show the
patient’s drawing when that stimulus was presented. During this experiment,
different patterns and single electrode stimulation were evaluated. The red circle
represents the location of the fovea

3.6 Conclusions
Based on the experiments applied in these 5 patients we can observe that
their responses to direct electrical stimulation were very variable across patients.
Such variability could be attributed to other medical conditions besides retinitis
pigmentosa such as spontaneous retinal responses and Charles Bonnet syndrome.

167
In addition, we think that the time that patients have been implanted for
could also be a factor of introducing some variability in the results. For instance,
Subject number 5 showed the best response among all patients and even though he
was implanted with the Argus II system in 2015, he has been implanted with the
Argus I system since 2004. Being implanted for a longer period of time has allowed
this patient to be exposed to more training with the system which could help him
understand better what he is perceiving when the system is ON.

From our 5 patients, we presented the tactile targets to three of them. All
three patients were able to draw successfully the tactile targets presented to them
over different trials. It was important to understand how their lack of visual
feedback and being bling for over 10 years can affect their perception in shape and
size of different objects.

Our results show that phosphene shapes are variable from electrode to
electrode but reproducible from trial to trial. Phosphenes are observed in different
shapes including circular, oval, semicircles, lines and even squares. These results
are consisted with Devyani’s work where she mentions that her subjects reported
seeing thin lines, arcs, circles, ovals and other sharply drawn wedges.

Our initial hypothesis was that a patient perceiving more circular
phosphenes than elongated phosphenes was associated with their capability to
perceive more complex shapes like letters, but based on our results we found that
168
this is not necessarily true. Some patients with round perceptions during single
electrode stimulation were unsuccessful at recognizing letters or simple shapes.
More experimental testing will need to be done to have a stronger conclusion.

Since this is a project still in progress, results presented here are pilot
studies, and more testing will need to be done to have a strong conclusion about
what other factors can be affecting each of the patients response.

169
CHAPTER 5
Conclusions And Future Work

For the development of a high-resolution retinal prosthesis, some technical
issues will need to be resolved. In my thesis I discussed different experimental
techniques that will enable the improvement of the current technology and could
potentially be used in the development of a high-resolution artificial vision. We
focused on evaluating new experimental models to help find solutions to electrode-
retinal distance, electrode size and axonal stimulation problems.

Different approaches have been utilized to attach the retina prosthesis to
the retina as an alternative to a retinal tack. We developed a technique that allows
us to use disintegrin-integrin binding as a novel attachment method for retinal
tissue. We demonstrated a technical approach for functionalizing the surface of
silicone by lasing and producing an active area to promote adhesion of disintegrins,
a capability that could be very advantageous in future medical devices. All our
experiments were performed in rabbits and cadaveric pig eyes, but may be
applicable to brain cortex, spinal cord and different areas of the body where
sutures are not an option.

170
Reversible bioadhesives have multiple applications for different medical
devices, and this is why it is important to have a sterilization technique bioactive
material friendly. We tested a sterilization technique and proved to be reliable for
bioactive materials. This involved sterilization of the silicone device before lasing
using a clinical ethylene oxide protocol and then doing the lasing and disintegrin
attachment under sterile conditions.

Future experiments should focus on developing retinal arrays coated with
silicone to allow the lasing and attachment of disintegrins to the array. Modifying
and improving the array design will allow the testing of this reversible bioadhesive
that could possibly be the solution to removing the tack and minimizing focal
pressure effects.

The second technical issue addressed in this thesis is that the design of a
high resolution retinal implant requires fitting 600-1000 electrodes in a 5mm
diameter, which will require smaller electrodes than the ones currently used which
it also mean that higher charge densities will be required to have neural response.
It is unknown if these new charge density values are safe for the retinal tissue. We
developed a new method that allows us to study the effects of electrical stimulation
on the retina in an in vivo model during the stimulation period using and Optical
Coherence Tomography (OCT). This new technique allowed us to study the long-
term effects of retinal stimulation.

171
Electrical stimulation at very high charge densities causes swelling of the
retina within minutes of stimulation, showing that frequency, charge density and
distance play an important role in the change of retinal thickness during
stimulation. OCT images showed swelling of within several retinal layers. The
stimulus levels at which swelling was noted is significantly higher than parameters
used generally for stimulation in humans in terms of rate and charge density.

It is not clear why electrical stimulation causes damage to the tissue, so
further analysis need to be performed to better understand the morphological
changes caused by the electrical stimulation. Currently retinal tissue acquired
during our experiments is being processed and stained with H&E and
immunofluorescence. Hematoxylin and Eosin (H&E) shows morphological changes
in retinal layers, GFAP is the best marker reflecting altered glial reactivity in the
retina; NeuN is a marker for ganglion cells and the ganglion cell layer and Iba1is a
calcium-binding protein highly expressed in microglial cells.

We also believe that it would be very important for the field to create a
mathematical model that will help find the safe and unsafe electrical stimulation
parameters for neural tissue. In addition to testing more combinations of
frequency, charge density and distance, future studies should explore the effects
that variations in electrode size and duty cycle have on tissue damage.

172
Lastly, in our third chapter, we studied the visualization of elongated
percepts in patients implanted with the Argus II system when one single electrode
was being stimulated. Shape perception is important to generate useful vision in
blind patients. We showed that stimulation with short duration pulses (0.45 ms)
produces percepts that are variable in size, and shape but reproducible from trial
to trial. The perceptual experience from particular stimulation parameters differs
across subjects

Our initial hypothesis was that a patient perceiving more circular
phosphenes than elongated phosphenes was associated with their capability to
perceive more complex shapes like letters, but based on our results we found that
this is not necessarily true. Some patients with round perceptions during single
electrode stimulation were unsuccessful at recognizing letters or simple shapes.
More experimental testing will need to be done to have a stronger conclusion.

A new experimental Video Processing Unit (VPU) has been developed by
Second Sight Medical Products that will allow us to perform different protocols
with patients. Extensive research has been done in variation of frequency and
amplitude of the stimulation parameters, but pulse width variation is yet to be
explored. Future experiments will focus on studying how changes in pulse width
will affect the phosphene perception to each patient.

173
References

[1] J. V Forrester, A. D. Dick, P. G. McMenamin, F. Roberts, and E. Pearlman, The
Eye Basic sciences in practice, Fourth. Elsevier Ltd, 2016.
[2] E. R. Kandel, J. H. Schwartz, and T. M. Jessell, Principles of neural science,
Fourth. McGraw- Hill, 2000.
[3] A. Lens, S. Coyne, Ledford, Janice, and Nemeth, “Ocular Anatomy and
Physiology Jump- Start,” in Ocular Anatomy and Physiology, no. January,
SLACK Incorporated, 2008, pp. 1–8.
[4] H. Wässle, “PARALLEL PROCESSING IN THE MAMMALIAN RETINA,” Nat.
Rev., vol. 5, no. October, pp. 1–11, 2004.
[5] J. C. Besharse and D. Bok, The retina and its disorders. Elsevier Ltd, 2014.
[6] S. C. Nemeth, C. Shea, M. Disclafani, and M. Schluter, “The posterior segment,”
in Ocular Anatomy and Physiology, 2nd ed., no. May, SLACK Incorporated,
2016, pp. 84–106.
[7] J. C. Besharse and D. Bok, The retina and its disorders. Elsevier Ltd, 2011.
[8] A. Reichenbach and A. Bringmann, “New Functions of Muller cells,” Glia, vol.
61, no. 5, pp. 651–678, 2013.
[9] A. Bringmann, I. Iandiev, T. Pannicke, A. Wurm, M. Hollborn, P. Wiedemann,
N. N. Osborne, and A. Reichenbach, “Cellular signaling and factors involved in
Muller cell gliosis: Neuroprotective and detrimental effects,” Prog. Retin. Eye
Res., vol. 28, no. 6, pp. 423–451, 2009.
174
[10] M. Kaneda, “Signal processing in the mammalian retina,” J Nippon Med Sch,
vol. 80, no. 1, pp. 16–24, 2013.
[11] G. S. Mannu, “Retinal phototransduction,” Neurosciences, vol. 19, no. 4, pp.
275–280, 2014.
[12] E. Kerr, “Back to basics : aged-related macular degeneration,” Nurs. Resid.
Care, vol. 15, no. 7, 2013.
[13] H. ur Rehman, “Retinitis Pigmentosa,” N. Z. Med. J., vol. 128, no. 1415, pp. 54–
56, 2015.
[14] A. C. Ho and C. D. Regillo, Aged-related macular degeneration Diagnosis and
treatment. Springer, 2011.
[15] M. A. Musarella and I. M. Macdonald, “Current Concepts in the Treatment of
Retinitis Pigmentosa,” J. Ophthalmol., vol. 2011, pp. 1–9, 2010.
[16] A. Santos, M. S. Humayun, E. De Juan, M. J. Marsh, I. B. Klock, and A. H. Milam,
“Preservation of the inner retina in retinitis pigmentosa,” Arch. Ophthalmol.,
vol. 115, no. 4, 1997.
[17] S. Y. Kim, S. Sadda, M. S. Humayun, and E. D. E. Juan, “MORPHOMETRIC
ANALYSIS OF THE MACULA IN EYES WITH GEOGRAPHIC ATROPHY DUE TO
AGE-RELATED MACULAR DEGENERATION,” Retina, vol. 22, pp. 464–470,
2002.
[18] O. Foerster, “Beitriige zur Pathophysiologie der Sehbahn und der Sehsphare,”
J. Psychol. Neurol, vol. 39, pp. 463–485, 1929.
[19] F. KRAUSE and H. SCHUM, “Neue deutsche Chirurgie,” KUTTNER, H., vol. 49a,
pp. 482–486, 1931.
175
[20] W. PENFIELD and T. RASMUSSEN, “The Cerebral Cortex ofMan.,” Macmillan,
1952.
[21] W. PENFIELD and H. JASPER, “Epilepsy and the Functional Anatomy of the
Human Brain,” London:Churchill, 1954.
[22] G. S. Brindley and W. S. Lewin, “The sensations produced by electrical
stimulation of the visual cortex,” J. Physiol., vol. 196, no. 2, pp. 479–493, 1968.
[23] G. S. Brindley and D. Rushton, “Implanted stimulators of the visual cortex as
visual prosthetic devices,” Trans. Am. Acad. Ophthalmol. Otolaryngol., vol. 78,
pp. 741–745, 1974.
[24] W. . Dobelle, M. . Mladejovsky, and J. . Girvin, “Artificial Vision for the Blind :
Electrical Stimulation of Visual Cortex Offers Hope for a Functional
Prosthesis,” Science (80-. )., vol. 183, no. 4123, pp. 440–444, 1974.
[25] M. Bak, J. . Girvin, F. . Hambrecht, C. . Kufta, G. . Loeb, and E. . Schmidt, “Visual
Sensations produced by intracortical microstimulation of the human
occipital cortex,” MEd. Biol. Eng. Comput., vol. 28, pp. 257–259, 1990.
[26] Z. Yang, Neural Computation , Neural Devices , and Neural Prosthesis, First.
Springer, 2014.
[27] T. Lin, H. Chang, C. Hsu, K. Hung, Y. Chen, S. Chen, and S. Chen, “Retinal
prostheses in degenerative retinal diseases,” J. Chinese Med. Assoc., vol. 78, no.
9, pp. 501–505, 2015.
[28] G. A. Williams, “Bimanual Technique for Retinal Tacking of Epiretinal
Prosthesis,” Retin. J. Retin. Vitr. Dis., vol. 36, no. 1, pp. 199–202, 2016.
[29] N. L. Opie, A. N. Burkitt, S. Member, H. Meffin, and D. B. Grayden, “Heating of
176
the Eye by a Retinal Prosthesis : Modeling , Cadaver and In Vivo Study,” IEEE
Trans. Biomed. Eng., vol. 59, no. 2, pp. 339–345, 2012.
[30] M. E. Çelik and İ. Karagö, “The Effect of the Electrical Stimulation on
Temperature Rise in the Retinal Tissue for Visual Prostheses,” Int. J. Comput.
Electr. Eng., vol. 6, no. 4, pp. 369–372, 2014.
[31] L. Yue, J. D. Weiland, B. Roska, and M. S. Humayun, “Retinal stimulation
strategies to restore vision : Fundamentals and systems,” Prog. Retin. Eye
Res., vol. 53, pp. 21–47, 2016.
[32] J. D. Weiland, W. Liu, and M. S. Humayun, “RETINAL PROSTHESIS,” pp. 361–
401, 2005.
[33] J. D. Weiland, S. Member, and M. S. Humayun, “Retinal Prosthesis,” IEEE
Trans. Biomed. Eng., vol. 61, no. 5, pp. 1412–1424, 2014.
[34] D. Yanai, J. D. Weiland, M. Mahadevappa, R. J. Greenberg, I. Fine, and M. S.
Humayun, “Visual Performance Using a Retinal Prosthesis in Three Subjects
With Retinitis Pigmentosa,” Am. J. Ophthalmol., vol. 143, no. 5, pp. 820–827,
2007.
[35] L. Yue, P. Falabella, P. Christopher, V. Wuyyuru, J. Dorn, P. Schor, R. J.
Greenberg, J. D. Weiland, and M. S. Humayun, “Ten-Year Follow-up of a Blind
Patient Chronically Implanted with Epiretinal Prosthesis Argus I,”
Ophthalmology, vol. 122, no. 12, pp. 2545–2552, 2015.
[36] Y. H.-L. Luo and L. da Cruz, “The Argus® II Retinal Prosthesis System,” Prog.
Retin. Eye Res., vol. 50, pp. 89–107, 2016.
[37] H. C. Stronks and G. Dagnelie, “The functional performance of the Argus II
177
retinal prosthesis.,” Expert Rev. Med. Devices, vol. 11, no. 1, pp. 23–30, 2014.
[38] M. S. Humayun, J. D. Dorn, L. Da Cruz, G. Dagnelie, J. A. Sahel, P. E. Stanga, A. V.
Cideciyan, J. L. Duncan, D. Eliott, E. Filley, A. C. Ho, A. Santos, A. B. Safran, A.
Arditi, L. V. Del Priore, and R. J. Greenberg, “Interim results from the
international trial of second sight’s visual prosthesis,” Ophthalmology, vol.
119, no. 4, pp. 779–788, 2012.
[39] G. Dagnelie, P. Christopher, A. Arditi, L. Cruz, J. L. Duncan, A. C. Ho, L. C. O. De
Koo, J. Sahel, P. E. Stanga, G. Thumann, Y. Wang, M. Arsiero, J. D. Dorn, R. J.
Greenberg, and I. I. Study, “Performance of real-world functional vision tasks
by blind subjects improves after implantation with the Argus ® II retinal
prosthesis system,” Clin. Exp. Ophthalmol., no. 5, 2016.
[40] A. K. Ahuja, J. Yeoh, J. D. Dorn, A. Caspi, V. Wuyyuru, M. J. Mcmahon, M. S.
Humayun, R. J. Greenberg, A. Ii, and S. Group, “Factors Affecting Perceptual
Threshold in Argus II Retinal Prosthesis Subjects,” vol. 2, no. 4, 2013.
[41] G. Roessler, T. Laube, C. Brockmann, T. Kirschkamp, B. Mazinani, M. Goertz, C.
Koch, I. Krisch, B. Sellhaus, H. K. Trieu, J. Weis, N. Bornfeld, H. Ro, A. Messner,
W. Mokwa, and P. Walter, “Implantation and Explantation of a Wireless
Epiretinal Retina Implant Device : Observations during the EPIRET3
Prospective Clinical Trial,” IOVS, vol. 50, no. 6, pp. 3003–3008, 2009.
[42] C. Koch, W. Mokwa, P. Walter, and M. Goertz, “First Results of a Study on a
Completely Implanted Retinal Prosthesis in Blind Humans,” IEEE Sensors
Conf., pp. 1237–1240, 2008.
[43] J. Menzel- Severing, T. Laube, C. Brockmann, N. Bornfeld, W. Mokwa, B.
178
Mazinani, P. Walter, and G. Roessler, “Implantation and explantation of an
active epiretinal visual prosthesis : 2-year follow-up data from the EPIRET3
prospective clinical trial,” Nat. Eye, vol. 26, no. 4, pp. 501–509, 2012.
[44] D. Hodgins, A. Bertsch, N. Post, M. Frischholz, B. Volckaerts, J. Spensley, J. M.
Wasikiewicz, H. Higgins, F. Von Stetten, and L. Kenney, “Healthy Aims:
Developing New Medical Implants and Diagnostic Equipment,” IEEE Comput.
Sci., pp. 14–21, 2008.
[45] T. Guenther, N. H. Lovell, and G. J. Suaning, “Bionic vision : system
architectures – a review,” Expert Rev. Med. Devices, vol. 9, no. 1, pp. 33–48,
2012.
[46] R. Hornig, T. Zehnder, M. Velikay-parel, T. Laube, M. Feucht, and G. Richard,
“The IMI Retinal Implant System,” in Artificial Sight, Springer, 2007, pp. 111–
128.
[47] G. Richard, R. Hornig, and M. Keseru, “Chronic Epiretinal Chip Implant in
Blind Patients with Retinitis Pigmentosa: Long Term Clinical Results,”
Investig. Ophthalmol. Vis. Sci., vol. 48, no. 13, p. 666, 2007.
[48] B. Gilly and K. Ishaque, “Pixium Vision.” [Online]. Available:
http://www.pixium-vision.com/en. [Accessed: 10-Apr-2016].
[49] D. Ghezzi, “Retinal prostheses : progress toward the next generation
implants,” Front. Neurosci., vol. 9, no. 290, pp. 1–6, 2015.
[50] D. Palanker, A. Vankov, P. Huie, and S. Baccus, “Design of a high-resolution
optoelectronic retinal prosthesis.,” J. Neural Eng., vol. 2, no. 1, pp. S105-20,
2005.
179
[51] A. Y. Chow, V. Y. Chow, K. H. Packo, J. S. Pollack, G. A. Peyman, and R.
Schuchard, “The Artificial Silicon Retina Microchip for the Treatment of
Vision Loss From Retinitis Pigmentosa,” Arch. Ophthalmol., vol. 122, pp. 460–
469, 2004.
[52] A. Y. Chow, A. K. Bittner, and M. T. Pardue, “The artificial silicon retina in
retinitis pigmentosa patients (an american ophthalmological association
thesis),” Trans Am Ophthalmol Soc, vol. 108, pp. 120–154, 2010.
[53] H. Lorach, O. Marre, J. Sahel, R. Benosman, and S. Picaud, “Neural stimulation
for visual rehabilitation : Advances and challenges,” J. Physiol. - Paris, vol.
107, no. 5, pp. 421–431, 2013.
[54] A. Y. Chow, “Retinal Prostheses Development in Retinitis Pigmentosa Patients
- Progress and Comparison,” Asia Pacific Acad. Ophthalmol., vol. 2, no. 4, pp.
253–268, 2013.
[55] A. T. Chuang, C. E. Margo, and P. B. Greenberg, “Retinal implants : a
systematic review,” Br J Ophthalmol, vol. 98, pp. 852–856, 2014.
[56] D. Besch, H. Sachs, P. Szurman, D. Gu, R. Wilke, S. Reinert, E. Zrenner, and F.
Gekeler, “Extraocular surgery for implantation of an active subretinal visual
prosthesis with external connections : feasibility and outcome in seven
patients,” Br J Ophthalmol, vol. 92, pp. 1361–1368, 2008.
[57] R. Wilke, V. Gabel, H. Sachs, K. B. Schmidt, F. Gekeler, D. Besch, P. Szurman, A.
Stett, B. Wilhelm, T. Peters, A. Harscher, U. Greppmaier, S. Kibbel, and H.
Benav, “Spatial Resolution and Perception of Patterns Mediated by a
Subretinal 16-Electrode Array in Patients Blinded by Hereditary Retinal
180
Dystrophies,” IOVS, vol. 52, pp. 5995–6003, 2011.
[58] E. Zrenner, K. U. Bartz-schmidt, H. Benav, D. Besch, A. Bruckmann, V. Gabel, F.
Gekeler, U. Greppmaier, A. Harscher, S. Kibbel, J. Koch, A. Kusnyerik, T. Peters,
K. Stingl, A. Stett, P. Szurman, B. Wilhelm, R. Wilke, E. Zrenner, K. Ulrich, B.
Schmidt, H. Benav, D. Besch, A. Bruckmann, V. Gabel, F. Gekeler, U.
Greppmaier, A. Harscher, S. Kibbel, J. Koch, A. Kusnyerik, T. Peters, K. Stingi,
H. Sachs, A. Stett, P. Szurman, B. Wilhehn, and R. Wilke, “Subretinal electronic
chips allow blind patients to read letters and combine them to words,” R.
Soc., vol. 278, pp. 1489–1497, 2011.
[59] K. Stingl, K. U. Bartz-schmidt, D. Besch, A. Braun, A. Bruckmann, F. Gekeler, U.
Greppmaier, C. Kernstock, A. Koitschev, S. Hipp, G. Ho, A. Kusnyerik, H. Sachs,
A. Schatz, K. T. Stingl, T. Peters, B. Wilhelm, and E. Zrenner, “Artificial vision
with wirelessly powered subretinal electronic implant alpha-IMS,” R. Soc., pp.
1–8, 2013.
[60] Z. M. Hafed, K. Stingl, K. Bartz-schmidt, F. Gekeler, and E. Zrenner,
“Oculomotor behavior of blind patients seeing with a subretinal visual
implant,” Vision Res., vol. 118, pp. 119–131, 2016.
[61] K. Stingl, K. U. Bartz-schmidt, D. Besch, C. K. Chee, C. L. Cottriall, F. Gekeler, M.
Groppe, T. L. Jackson, R. E. Maclaren, A. Koitschev, A. Kusnyerik, J. Neffendorf,
J. Nemeth, M. Adheem, N. Naeem, T. Peters, J. D. Ramsden, H. Sachs, A.
Simpson, M. S. Singh, B. Wilhelm, D. Wong, and E. Zrenner, “Subretinal Visual
Implant Alpha IMS – Clinical trial interim report,” Vision Res., vol. 111, pp.
149–160, 2015.
181
[62] J. F. Rizzo III, J. Wyatt, J. Loewenstein, S. Kelly, and D. Shire, “Perceptual
Efficacy of Electrical Stimulation of Human Retina with a Microelectrode
Array during Short-Term Surgical Trials,” IOVS, vol. 44, no. 12, pp. 5362–
5369, 2003.
[63] J. F. Rizzo III, “Update on Retinal Prosthetic Research: The Boston Retinal
Implant Project,” J. Neuro-Ophthalmology, vol. 31, pp. 160–168, 2011.
[64] S. K. Kelly, D. B. Shire, J. Chen, P. Doyle, M. D. Gingerich, S. F. Cogan, W. A.
Drohan, S. Behan, L. Theogarajan, J. L. Wyatt, and J. F. Rizzo III, “A Hermetic
Wireless Subretinal Neurostimulator for Vision Prostheses,” IEEE Trans.
Biomed. Eng., vol. 58, no. 11, pp. 3197–3205, 2011.
[65] S. K. Kelly, D. B. Shire, J. Chen, D. Marcus, S. F. Cogan, W. A. Drohan, W.
Ellersick, A. Krishnan, S. Behan, J. L. Wyatt, and J. F. Rizzo III, “Developments
on the Boston 256-channel retinal implant,” IEEE Int. Conf. Multimed. Expo
Work., pp. 7–12, 2013.
[66] D. Palanker, A. Vankov, P. Huie, and S. Baccus, “Design of a high-resolution
optoelectronic retinal prosthesis,” J. Neural Eng., vol. 2, pp. 105–120, 2005.
[67] K. Mathieson, J. Loudin, G. Goetz, P. Huie, L. Wang, T. I. Kamins, L. Galambos,
R. Smith, J. S. Harris, A. Sher, and D. Palanker, “photovoltaic retinal prosthesis
with pixel density,” Nat. Photonics, vol. 6, no. June, pp. 391–398, 2012.
[68] Y. Mandel, G. Goetz, D. Lavinsky, P. Huie, K. Mathieson, L. Wang, T. Kamins, L.
Galambos, R. Manivanh, J. Harris, and D. Palanker, “Cortical responses elicited
by photovoltaic subretinal prostheses exhibit similarities to visually evoked
potentials,” Nat. Commun., vol. 4, no. 1980, pp. 1–9, 2013.
182
[69] H. Lorach, G. Goetz, Y. Mandel, X. Lei, T. I. Kamins, K. Mathieson, P. Huie, R.
Dalal, J. S. Harris, and D. Palanker, “Performance of photovoltaic arrays in-
vivo and characteristics of prosthetic vision in animals with retinal
degeneration,” Vision Res., vol. 111, pp. 142–148, 2015.
[70] D. Y. Lee, H. Lorach, P. Huie, and D. Palanker, “Implantation of Modular
Photovoltaic Subretinal Prosthesis,” Ophthalmic Surg Imaging Lasers Retin.,
vol. 47, no. 2, pp. 171–175, 2016.
[71] L. Schmetterer and J. W. Kiel, Ocular Blood Flow. Springer, 2009.
[72] Y. Yan, X. Sui, W. Liu, Y. Lu, P. Cao, Z. Ma, Y. Chen, X. Chai, and L. Li, “Spatial
characteristics of evoked potentials elicited by a MEMS microelectrode array
for suprachoroidal-transretinal stimulation in a rabbit,” Graefe’s Arch. Clin.
Exp. Ophthalmol., vol. 253, no. 9, pp. 1515–1528, 2015.
[73] P. M. Lewis, L. N. Ayton, R. H. Guymer, A. J. Lowery, P. J. Blamey, P. J. Allen, D.
Luu, and J. V Rosenfeld, “Advances in implantable bionic devices for
blindness : a review,” R. Australas. Coll. Surg., pp. 1–6, 2016.
[74] N. Barnes, A. F. Scott, P. Lieby, M. A. Petoe, C. Mccarthy, A. Stacey, L. N. Ayton,
N. C. Sinclair, M. N. Shivdasani, N. H. Lovell, H. J. Mcdermott, and J. G. Walker,
“Vision function testing for a suprachoroidal retinal prosthesis: effects of
image filtering,” J. Neural Eng., vol. 13, no. 3, pp. 1–15, 2016.
[75] L. N. Ayton, P. J. Blamey, R. H. Guymer, and C. D. Luu, “First-in-Human Trial of
a Novel Suprachoroidal Retinal Prosthesis,” PLoS One, vol. 9, no. 12, pp. 1–26,
2014.
[76] M. N. Shivdasani, N. C. Sinclair, P. N. Dimitrov, M. Varsamidis, L. N. Ayton, C. D.
183
Luu, T. Perera, H. J. Mcdermott, and P. J. Blamey, “Factors Affecting
Perceptual Thresholds in a Suprachoroidal Retinal Prosthesis,” IOVS, vol. 55,
no. 10, pp. 6467–6481, 2014.
[77] T. Fujikado, M. Kamei, H. Sakaguchi, H. Kanda, T. Morimoto, Y. Ikuno, K.
Nishida, and H. Kishima, “Testing of Semichronically Implanted Retinal
Prosthesis by Suprachoroidal-Transretinal Stimulation in Patients with
Retinitis Pigmentosa,” IOVS, vol. 52, no. 7, pp. 4726–4733, 2011.
[78] T. Fujikado, M. Kamei, H. Kishima, T. Morimoto, H. Kanda, H. Sakaguchi, K.
Nishida, T. Endo, K. Osawa, and M. Ozawa, “One-Year Outcomes of 49-
Channel Suprachoroidal-Transretinal Stimulation (STS) Retinal Prosthesis in
Patients with Advanced Retinitis Pigmentosa,” in The Association for Research
in Vision and Ophthalmology, 2016, p. 5203.
[79] A. K. Ahuja and M. R. Behrend, “The Argus II retinal prosthesis: Factors
affecting patient selection for implantation,” Prog. Retin. Eye Res., vol. 36, pp.
1–23, 2013.
[80] D. Nanduri, I. Fine, A. Horsager, G. M. Boynton, M. S. Humayun, R. J.
Greenberg, and J. D. Weiland, “Frequency and amplitude modulation have
different effects on the percepts elicited by retinal stimulation,” Investig.
Ophthalmol. Vis. Sci., vol. 53, no. 1, pp. 205–214, 2012.
[81] D. Nanduri, “Prosthetic vision in blind human patients: Predicting the
percepts of epiretinal stimulation,” University of Southern California, 2011.
[82] A. C. Weitz, “Improving stimulation strategies for epiretinal prostheses,”
University of Southern California, 2013.
184
[83] C. De Balthasar, S. Patel, A. Roy, R. Freda, S. Greenwald, A. Horsager, M.
Mahadevappa, D. Yanai, M. J. McMahon, M. S. Humayun, R. J. Greenberg, J. D.
Weiland, and I. Fine, “Factors affecting perceptual thresholds in epiretinal
prostheses,” Investig. Ophthalmol. Vis. Sci., vol. 49, no. 6, pp. 2303–2314,
2008.
[84] C. Sekirnjak, P. Hottowy, A. Sher, W. Dabrowski, A. M. Litke, and E. J.
Chichilnisky, “Electrical Stimulation of Mammalian Retinal Ganglion Cells
With Multielectrode Arrays,” J. Neurophysiol., vol. 95, pp. 3311–3327, 2006.
[85] A. Fumitaka and K. Junko, “A Plastic Tack for the Treatment of Retinal
Detachment with Giant Tear,” Am. J. Ophthalmol., vol. 95, no. 2, pp. 260–261,
1983.
[86] E. De Juan Jr, B. W. McCuen II, and R. Machemer, “The Use of Retinal Tacks in
the Repair of Complicated Retinal Detachments,” Am. J. Ophthalmol., vol. 102,
no. 1, pp. 20–24, 1986.
[87] B. C. Basinger, “Modeling Retinal Prosthesis Mechanics,” University of
Southern California, 2009.
[88] D. Ivastinovic, G. Langmann, M. Asslaber, T. Georgi, A. Wedrich, and M.
Velikay-parel, “Distribution of glial fibrillary acidic protein accumulation
after retinal tack insertion for intraocular fixation of epiretinal implants,”
Acta Ophthalmol., pp. 416–417, 2012.
[89] M. Barczyk, S. Carracedo, and D. Gullberg, “Integrins,” Cell Tissue Res, vol. 339,
pp. 269–280, 2010.
[90] C. M. Helchowski, “MECHANISTIC STUDIES OF THE DISINTEGRIN
185
CONTORTROSTATIN AND CHARACTERIZATION OF THE RECOMBINANT
PROTEIN VICROSTATIN,” 2010.
[91] A. . Dupuy and E. Caron, “Integrin-dependent phagocytosis: spreading from
microadhesion to new concepts,” J. Cell Sci, vol. 121, no. 11, pp. 1773–83,
2008.
[92] R. . Hynes, “Integrins: versatility, modulation, and signaling in cell adhesion,”
Cell, vol. 69, no. 1, pp. 11–25, 1992.
[93] C. D.A, “Structural and biologic properties of integrin-mediated cell
adhesion,” Clin Lab Med, vol. 12, no. 2, pp. 217–36, 1992.
[94] S. Lucena, R. Castro, C. Lundin, A. Hofstetter, A. Alaniz, M. Suntravat, and E. E.
Sánchez, “Inhibition of pancreatic tumoral cells by snake venom
disintegrins,” Toxicon, vol. 93, pp. 136–143, 2015.
[95] J. A. Askari, P. A. Buckley, P. A. Mould, and M. J. Humphries, “Linking integrin
conformation to function,” J. Cell Sci., vol. 122, no. 2, pp. 165–170, 2009.
[96] L. Gillan, D. Matei, D. A. Fishman, C. S. Gerbin, B. Y. Karlan, and D. D. Chang,
“Periostin Secreted by Epithelial Ovarian Carcinoma Is a Ligand for ␣ V ␤ 3
and ␣ V ␤ 5 Integrins and Promotes Cell Motility 1,” Cancer Res., vol. 62, pp.
5358–5364, 2002.
[97] N. Ahmed, “Overexpression of alpha(v)beta6 integrin in serous epithelial
ovarian cancer regulates extracellular matrix degradation via the
plasminogen activation cascade,” Carcinogenesis, vol. 23, no. 2, pp. 237–244,
2002.
186
[98] C. Guidry, K. M. Bradley, and J. L. King, “Tractional force generation by human
muller cells: growth factor responsiveness and integrin receptor
involvement,” Invest Ophthalmol Vis Sci, vol. 44, no. 3, pp. 1355–1363, 2003.
[99] S. Swenson, S. Ramu, and F. S. Markland, “Anti-angiogenesis and-RGD-
containing snake venom disintegrins,” Curr. Pharm. Des., vol. 13, no. 28, pp.
2860–2871, 2007.
[100] M. A. Mclane, C. Marcinkiewicz, S. Vijay-Kumar, I. Wierzbicka-Patynowski,
and S. Niewiarowski, “Viper Venom Disintegrins and Related Molecules,” Soc.
Exp. Biol. Med., vol. 219, no. 2, pp. 109–119, 1998.
[101] M. Trikha, W. E. Rote, P. J. Manley, L. R. Benedict, and F. S. Markland,
“Purification and characterization of platelet aggregation inhibitors from
snake venoms,” Thromb. Res., vol. 73, no. 213, pp. 39–52, 1994.
[102] R. O. Minea, C. M. Helchowski, S. J. Zidovetzki, F. K. Costa, S. D. Swenson, and
F. S. Markland, “Vicrostatin - An anti-invasive multi-integrin targeting
chimeric disintegrin with tumor anti-angiogenic and pro-apoptotic
activities,” PLoS One, vol. 5, no. 6, 2010.
[103] Q. Zhou and P. Hu, “Molecular cloning and functional expression of
contortrostatin, a homodimeric disintegrin from southern copperhead snake
venom,” Arch Biochem Biophys, vol. 375, no. 2, pp. 278–88, 2000.
[104] A. Rowley, “THE ELASTIC PROPERTIES OF THE EYEWALL , THE EFFECT OF
FOCAL PRESSURE ON THE RETINA , AND THE DEVELOPMENT OF
REVERSIBLE BIOADHESIVES FOR UTILIZATION IN A RETINAL PROSTHESIS
by Adrian Paul Rowley A Dissertation Presented to the FACULTY OF THE USC
187
GRADUATE S,” no. August, 2011.
[105] A. Colas and J. Curtis, “Silicones,” in Handbook of Polymer Applications in
Medicine and Medical Devices, K. Modjarrad and S. Ebnesajjas, Eds. Chadds
Ford, PA: Elsevier Inc., 2013, pp. 131–143.
[106] F. S. M. J. Adrian P. Rowley, :ucien D. Laude, Mark S. Humayun, James D.
Weiland, Atoosa Lotfi, “Biocompatible Implants and Methods of Making and
Attaching the Same,” 2015.
[107] A. P. Rowley, L. D. Laude, M. S. Humayun, J. D. Weiland, A. Lofti, and F. S.
Markland, “Biocompatible implants and methods of making and attaching the
same.pdf,” US20150105864, 2015.
[108] A. . Rowley, L. Laude, A. Lofti, K. Kolev, S. Swenson, F. Markland, J. D. Weiland,
and M. S. Humayun, “Reversibly interfacing biomaterials with the retina,”
Investig. Ophthalmol. Vis. Sci., vol. 51, no. 13, p. 3041, 2010.
[109] I. Donati, M. Benincasa, M. Foulc, G. Turco, M. Toppazzini, D. Solinas, S.
Spilimbergo, I. Kikic, and S. Paoletti, “Terminal Sterilization of BisGMA-
TEGDMA Thermoset Materials and Their Bioactive Surfaces by Supercritical
CO 2,” Biomacromolecules, vol. 13, pp. 1152–1160, 2012.
[110] K. Fox, H. Meffin, O. Burns, C. J. Abbott, P. J. Allen, N. L. Opie, C. Mcgowan, J.
Yeoh, A. Ahnood, C. D. Luu, R. Cicione, A. L. Saunders, M. Mcphedran, L.
Cardamone, J. Villalobos, D. J. Garrett, D. A. X. Nayagam, N. V Apollo, K.
Ganesan, M. N. Shivdasani, A. Stacey, M. Escudie, S. Lichter, R. K. Shepherd, S.
Prawer, and F. O. X. E. T. Al, “Development of a Magnetic Attachment Method
for Bionic Eye Applications,” Artif. Organs, vol. 40, no. 3, pp. E12–E24, 2016.
188
[111] E. Margalit, G. Y. Fujil, J. C. Lai, P. Gupta, S.-J. Chen, J.-S. Shyu, D. V Piyathaisere,
J. D. Weiland, E. J. De Juan, and M. S. Humayun, “Bioadhesives for intraocular
use,” Retina, vol. 20, pp. 469–477, 2000.
[112] V. V Khutoryanskiy, Mucoadhesive Materials and Drug delivery Systems. Wiley,
2014.
[113] A. L. Hodgkin and A. F. Huxley, “A Quantitative Description of Membrane
Current and its Application to Conduction and Excitation in Nerve,” J. Physiol.,
vol. 117, pp. 500–544, 1952.
[114] M. Van Dongen and W. Serdijn, Design of Efficient and Safe Neural Stimulators,
First. Springer, 2016.
[115] A. Ray, “Effect of continuous electrical stimulation on retinal structure and
function,” PhD Propos., vol. 1, no. August, 2015.
[116] L.-H. Chan, “Electrical excitation of degenerate retina,” University of Southern
California, 2009.
[117] D. T. Brocker and W. M. Grill, “Principles of electrical stimulation of neural
tissue,” in Handbook of Clinical Neurology. Brain Stimulation, 3rd ed., vol. 116,
A. . Lozano and M. Hallet, Eds. Elsevier B.V., 2013, pp. 3–18.
[118] N. J. . Rijkhoff, J. Holsheimer, E. . Koldewijin, J. . Struijk, P. E. . van Kerrebroeck,
F. M. . Debruyne, and H. Wijkstra, “Selective stimulation of sacral nerve roots
for bladder control: A study by computer modeling,” IEEE Trans. Biomed.
Eng., vol. 41, no. 5, pp. 413–424, 1994.
[119] D. Zhou and E. Greenbaum, Implantable Neural Prostheses 1 Devices and
Applications. Springer, 2009.
189
[120] D. D. Zhou and E. Greenbaun, Implantable Neural Prostheses 2. Springer,
2010.
[121] H. Kaim, A. Rothermel, and N. Pour Aryan, Stimulation and Recording
Electrodes for Neural Prostheses. 2015.
[122] E. Slavcheva, R. Vitushinsky, W. Mokwa, U. Schnakenberg, and E. Slavcheva,
“Sputtered Iridium Oxide Films as Charge Injection Material for Functional
Electrostimulation,” J. Electrochem. Soc., vol. 151, no. 7, pp. 226–237, 2004.
[123] W. D. Sproul, D. J. Christie, and D. C. Carter, “Control of reactive sputtering
processes,” Thin Solid Films, vol. 491, no. 1, pp. 1–17, 2005.
[124] A. Petrossians, J. W. Iii, J. D. Weiland, and F. Mansfeld, “Electrodeposition and
Characterization of Thin-Film Platinum-Iridium Alloys for Biological
Interfaces,” vol. 158, no. 5, pp. 269–276, 2011.
[125] L. Colodetti, J. D. Weiland, S. Colodetti, A. Ray, M. J. Seiler, D. R. Hinton, and M.
S. Humayun, “Pathology of damaging electrical stimulation in the retina,” Exp.
Eye Res., vol. 85, no. 1, pp. 23–33, 2007.
[126] D. R. Merrill, M. Bikson, and J. G. R. Jefferys, “Electrical stimulation of
excitable tissue : design of efficacious and safe protocols,” J. Neurosci.
Methods, vol. 141, pp. 171–198, 2005.
[127] W. S. Meaking, J. Edgerton, C. W. Wharton, and R. A. Meldrum,
“Electroporation-induced damage in mammalian cell DNA,” Biochim. Biophys.
Acta, vol. 1264, no. 3, pp. 357–362, 1995.
[128] L. P. Gartner and J. L. Hiatt, Texto Atlas de histologia, Second. McGraw- Hill,
2001.
190
[129] S. J. Ryan and S. R. Sadda, Ryan’s retinal imaging and diagnostics, First. Los
Angeles: Elsevier, 2013.
[130] R. V. Shannon, “A Model of Safe Levels for Electrical Stimulation,” IEEE Trans.
Biomed. Eng., vol. 39, no. 4, pp. 424–426, 1992.
[131] A. Ray, E.-J. Lee, M. S. Humayun, and J. D. Weiland, “Continuous electrical
stimulation decreases retinal excitability but does not alter retinal
morphology,” J. Neural Eng., vol. 8, no. 4, p. 45003, 2011.
[132] E. Cohen, A. Agrawal, M. Connors, B. Hansen, H. Charkhkar, and J. Pfefer,
“Optical coherence tomography imaging of retinal damage in real time under
a stimulus electrode.”
[133] A. Butterwick, A. Vankov, P. Huie, Y. Freyvert, and D. Palanker, “Tissue
Damage by Pulsed Electrical Stimulation,” vol. 54, no. 12, pp. 2261–2267,
2007.
[134] M. S. Humayun, J. D. Weiland, G. Y. Fujii, R. Greenberg, R. Williamson, J. Little,
B. Mech, V. Cimmarusti, G. Van Boemel, G. Dagnelie, and E. De Juan, “Visual
perception in a blind subject with a chronic microelectronic retinal
prosthesis,” Vision Res., vol. 43, no. 24, pp. 2573–2581, 2003.
[135] A. Horsager, S. H. Greenwald, J. D. Weiland, M. S. Humayun, R. J. Greenberg, M.
J. McMahon, G. M. Boynton, and I. Fine, “Predicting visual sensitivity in retinal
prosthesis patients,” Investig. Ophthalmol. Vis. Sci., vol. 50, no. 4, pp. 1483–
1491, 2009.
[136] W. F. Agnew and D. B. McCreery, “Considerations for safety with chronically
implanted nerve electrodes.,” Epilepsia, vol. 31 Suppl 2, pp. S27–S32, 1990.
191
[137] R. J. Jensen, O. R. Ziv, and J. F. Rizzo, “Responses of rabbit retinal ganglion
cells to electrical stimulation with an epiretinal electrode,” J. Neural Eng., vol.
2, no. 1, pp. s16–s21, 2005.
[138] L. H. Jepson, P. Hottowy, G. A. Weiner, W. Dabrowski, A. M. Litke, and E. J.
Chichilnisky, “High-fidelity reproduction of spatiotemporal visual signals for
retinal prosthesis,” Neuron, vol. 83, no. 1, pp. 87–92, 2014.
[139] H. A. Shah, S. R. Montezuma, and J. F. Rizzo, “In vivo electrical stimulation of
rabbit retina: Effect of stimulus duration and electrical field orientation,” Exp.
Eye Res., vol. 83, no. 2, pp. 247–254, 2006.
[140] S. T. Walston, R. H. Chow, and J. D. Weiland, “Patch clamp recordings of
retinal bipolar cells in response to extracellular electrical stimulation in
wholemount mouse retina,” Proc. Annu. Int. Conf. IEEE Eng. Med. Biol. Soc.
EMBS, vol. 2015, pp. 3363–3366, 2015.
[141] A. C. Weitz, D. Nanduri, M. R. Behrend, A. Gonzalez-Calle, R. J. Greenberg, M. S.
Humayun, R. H. Chow, and J. D. Weiland, “Improving the spatial resolution of
epiretinal implants by increasing stimulus pulse duration,” Sci. Transl. Med.,
vol. 7, no. 318, p. 318ra203-318ra203, 2015.
[142] A. K. Cho, “Understanding the degenerate retina’s response to electrical
stimulation: an in vitro approach,” University of Southern California, 2014.
[143] A. C. Weitz, M. R. Behrend, N. S. Lee, R. L. Klein, V. A. Chiodo, W. W. Hauswirth,
M. S. Humayun, J. D. Weiland, and R. H. Chow, “Imaging the response of the
retina to electrical stimulation with genetically encoded calcium indicators,”
J. Neurophysiol., vol. 109, no. 7, pp. 1979–1988, 2013.

Asset Metadata

Creator Gonzalez Calle, Alejandra (author)

Core Title Towards a high resolution retinal implant

Contributor Electronically uploaded by the author (provenance)

School Andrew and Erna Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Biomedical Engineering

Publication Date 02/15/2018

Defense Date 11/21/2017

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aged related macular degeneration,AMD,bioactive materials,biocompatibility,bipolar cells,blindness,clinical studies,disintegrins,electrical stimulation,electrode-tissue interface,epiretinal prosthesis,eye,ganglion cells,high charge density stimulation,high resolution retinal implant,integrins,OAI-PMH Harvest,retina,retinal prosthesis,retinal stimulation,retinal tack,retinitis pigmentosa,stimulation paradigms,stimulation safety

Language English

Advisor D'Argenio, David (committee chair), Humayun, Mark (committee member), Kashani, Amir (committee member), Weiland, James (committee member)

Creator Email gonz762@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-476355

Unique identifier UC11266652

Identifier etd-GonzalezCa-6046.pdf (filename),usctheses-c40-476355 (legacy record id)

Legacy Identifier etd-GonzalezCa-6046.pdf

Dmrecord 476355

Document Type Dissertation

Rights Gonzalez Calle, Alejandra

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Abstract (if available)

Abstract Several incurable diseases result in blindness for hundreds of thousands of individuals each year. Aged related macular degeneration (AMD) and Retinitis pigmentosa (RP) are two degenerative diseases that affect the photoreceptor cells causing the retina to lose its ability to translate light into electrical signals. RP is caused by mutations in over 100 genes that result in the degeneration of rod and cone receptors and the retinal pigmented epithelium (RPE). Patients will lose their night vision due to the degeneration of the rods, and their central vision in the later stages of the disease when macular cones start to degenerate. The overall incidence of retinitis pigmentosa is 1 in 4000 live births worldwide. AMD causes loss of vision in the center of the visual field. AMD is the most common cause of irreversible blindness in elderly individuals worldwide. Studies have shown that 30 percent of the ganglion cells and 78 percent of the inner nuclear layer cells remain largely intact in retinas with severe RP and AMD, thus allowing the retina to be electrically stimulated to restore a sense of vision. ❧ Retinal prostheses are being studied as a method to restore vision. They have demonstrated the capability to elicit the sensation of light and to give subjects more independence in their day to day activities. The results are remarkable because RP patients before implantation had bare or no light perception. Patients implanted with the Argus II system, in their 5 year follow up are able to perform tasks such as following a white line on the floor and finding a door in a room and walking towards it. ❧ Some technical issues remain unresolved, clinical studies of epiretinal implants reveal limitations in the ability of patients to determine the orientation of grating (used to measure visual acuity), and those who can recognize letters take more than 40 s to do so. High resolution retinal prostheses will require close proximity between the electrode array and the retina to maintain low stimulus thresholds and to minimize the power required by the system, smaller electrodes to accomplish precise activation of retinal cells to elicit a small visual phosphenes that can serve as a building block for the pattern, and new stimulation paradigms that will provide better control over the spatial patterns of activation. ❧ Currently a custom tack is used to attach the electrode array to the retina making it a challenge to control the electrode-retina distance overtime. During this report we study and present a novel attachment method based on disintegrin-integrin binding for retinal tissue. We developed a reliable technique to lase and produce an active area on the silicone surface, we found a reliable sterilization method for bioactive materials and we developed a new and reversible method of attachment to retinal tissue. ❧ Human subject testing has shown patients often see large, elongated phosphenes due to stimulation of RGC axon bundles. Smaller electrodes will allow more focal stimulation but will also require higher charge densities to have the same neural activation. It is unclear if these new high charge densities are safe for the tissue and this is why we developed a new method that allowed us to study the effects of electrical stimulation on the retina in an in vivo model during the stimulation period. This is important because it allowed us to study the retinal response in its natural conditions. In addition, this technique allowed us to study the long-term effects of retinal stimulation during survival experiments. We tested the safety of high charge density stimulation in the retina, which is a main factor that will need to be used in the future during the design of a high-resolution retinal prosthesis.