3D electronic implants in subretinal space: Long-term follow-up in rodents.

Clinical results with photovoltaic subretinal prosthesis (PRIMA) demonstrated restoration of sight via electrical stimulation of the interneurons in degenerated retina, with resolution matching the 100 µm pixel size. Since scaling the pixels below 75 µm in the current bipolar planar geometry will significantly limit the penetration depth of the electric field and increase stimulation threshold, we explore the possibility of using smaller pixels based on a novel 3-dimensional honeycomb-shaped design. We assessed the long-term biocompatibility and stability of these arrays in rats by investigating the anatomical integration of the retina with flat and 3D implants and response to electrical stimulation over lifetime - up to 32-36 weeks post-implantation in aged rats. With both flat and 3D implants, signals elicited in the visual cortex decreased after the day of implantation by more than 3-fold, and gradually recovered over the next 12-16 weeks. With 25 µm high honeycomb walls, the majority of bipolar cells migrate into the wells, while amacrine and ganglion cells remain above the cavities, which is essential for selective network-mediated stimulation of the retina. Retinal thickness and full-field stimulation threshold with 40 µm-wide honeycomb pixels were comparable to those with planar devices - 0.05 mW/mm2 with 10 ms pulses. However, fewer cells from the inner nuclear layer migrated into the 20 µm-wide wells, and stimulation threshold increased over 12-16 weeks, before stabilizing at about 0.08 mW/mm2. Such threshold is still significantly lower than 1.8 mW/mm2 with a previous design of flat bipolar pixels, confirming the promise of the 3D honeycomb-based approach to high resolution subretinal prosthesis.

Light-evoked deformations in rod photoreceptors, pigment epithelium and subretinal space revealed by prolonged and multilayered optoretinography.

Phototransduction involves changes in concentration of ions and other solutes within photoreceptors and in subretinal space, which affect osmotic pressure and the associated water flow. Corresponding expansion and contraction of cellular layers can be imaged using optoretinography (ORG), based on phase-resolved optical coherence tomography (OCT). Until now, ORG could reliably detect only photoisomerization and phototransduction in photoreceptors, primarily in cones under bright stimuli. Here, by employing a phase-restoring subpixel motion correction algorithm, which enables imaging of the nanometer-scale tissue dynamics during minute-long recordings, and unsupervised learning of spatiotemporal patterns, we discover optical signatures of the other retinal structures' response to visual stimuli. These include inner and outer segments of rod photoreceptors, retinal pigment epithelium, and subretinal space in general. The high sensitivity of our technique enables detection of the retinal responses to dim stimuli: down to 0.01% bleach level, corresponding to natural levels of scotopic illumination. We also demonstrate that with a single flash, the optoretinogram can map retinal responses across a 12° field of view, potentially replacing multifocal electroretinography. This technique expands the diagnostic capabilities and practical applicability of optoretinography, providing an alternative to electroretinography, while combining structural and functional retinal imaging in the same OCT machine.

Prosthetic Visual Acuity with the PRIMA Subretinal Microchip in Patients with Atrophic Age-Related Macular Degeneration at 4 Years Follow-up.

Objective: To assess the efficacy and safety of the PRIMA neurostimulation system with a subretinal microchip for improving visual acuity (VA) in patients with geographic atrophy (GA) due to age-related macular degeneration (AMD) at 48-months postimplantation. Design: Feasibility clinical trial of the PRIMA subretinal prosthesis in patients with atrophic AMD, measuring best-corrected ETDRS VA (Clinicaltrials.govNCT03333954). Subjects: Five patients with GA, no foveal light perception, and VA of logarithm of the minimum angle of resolution (logMAR) 1.3 to 1.7 (20/400-20/1000) in their worse-seeing "study" eye. Methods: In patients subretinally implanted with a photovoltaic neurostimulation array containing 378 pixels of 100 µm in size, the VA was measured with and without the PRIMA system using ETDRS charts at 1 m. The system's external components, augmented reality glasses, and pocket computer provide image processing capabilities, including zoom. Main Outcome Measures: Visual acuity using ETDRS charts with and without the system, as well as light sensitivity in the central visual field, measured by Octopus perimetry. Anatomical outcomes demonstrated by fundus photography and OCT up to 48 months postimplantation. Results: All 5 subjects met the primary end point of light perception elicited by the implant in the scotoma area. In 1 patient, the implant was incorrectly inserted into the choroid. One subject died 18 months postimplantation due to study-unrelated reasons. ETDRS VA results for the remaining 3 subjects are reported here. Without zoom, VA closely matched the pixel size of the implant: 1.17 ± 0.13 pixels, corresponding to a mean logMAR of 1.39, or Snellen of 20/500, ranging from 20/438 to 20/565. Using zoom at 48 months, subjects improved their VA by 32 ETDRS letters versus baseline (standard error 5.1) 95% confidence intervals (13.4, 49.9; P < 0.0001). Natural peripheral visual function in the treated eye did not decline after surgery or during the 48-month follow-up period (P = 0.08). Conclusions: Subretinal implantation of PRIMA in subjects with GA experiencing profound vision loss due to AMD is feasible and well tolerated, with no reduction of natural peripheral vision up to 48 months. Prosthetic central vision provided by photovoltaic neurostimulation enabled patients to reliably recognize letters and sequences of letters, and with zoom, it improved VA of up to 8 ETDRS lines. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Enhancing Prosthetic Vision by Upgrade of a Subretinal Photovoltaic Implant in situ.

In patients with atrophic age-related macular degeneration, subretinal photovoltaic implant (PRIMA) provided visual acuity up to 20/440, matching its 100µm pixels size. Next-generation implants with smaller pixels should significantly improve the acuity. This study in rats evaluates removal of a subretinal implant, replacement with a newer device, and the resulting grating acuity in-vivo. Six weeks after the initial implantation with planar and 3-dimensional devices, the retina was re-detached, and the devices were successfully removed. Histology demonstrated a preserved inner nuclear layer. Re-implantation of new devices into the same location demonstrated retinal re-attachment to a new implant. New devices with 22µm pixels increased the grating acuity from the 100µm capability of PRIMA implants to 28µm, reaching the limit of natural resolution in rats. Reimplanted devices exhibited the same stimulation threshold as for the first implantation of the same implants in a control group. This study demonstrates the feasibility of safely upgrading the subretinal photovoltaic implants to improve prosthetic visual acuity.

Three-dimensional electro-neural interfaces electroplated on subretinal prostheses.

Objective.Retinal prosthetics offer partial restoration of sight to patients blinded by retinal degenerative diseases through electrical stimulation of the remaining neurons. Decreasing the pixel size enables increasing prosthetic visual acuity, as demonstrated in animal models of retinal degeneration. However, scaling down the size of planar pixels is limited by the reduced penetration depth of the electric field in tissue. We investigated 3-dimensional (3d) structures on top of photovoltaic arrays for enhanced penetration of the electric field, permitting higher resolution implants.Approach.3D COMSOL models of subretinal photovoltaic arrays were developed to accurately quantify the electrodynamics during stimulation and verified through comparison to flat photovoltaic arrays. Models were applied to optimize the design of 3D electrode structures (pillars and honeycombs). Return electrodes on honeycomb walls vertically align the electric field with bipolar cells for optimal stimulation. Pillars elevate the active electrode, thus improving proximity to target neurons. The optimized 3D structures were electroplated onto existing flat subretinal prostheses.Main results.Simulations demonstrate that despite exposed conductive sidewalls, charge mostly flows via high-capacitance sputtered iridium oxide films topping the 3D structures. The 24µm height of honeycomb structures was optimized for integration with the inner nuclear layer cells in the rat retina, whilst 35µm tall pillars were optimized for penetrating the debris layer in human patients. Implantation of released 3D arrays demonstrates mechanical robustness, with histology demonstrating successful integration of 3D structures with the rat retinain-vivo.Significance. Electroplated 3D honeycomb structures produce vertically oriented electric fields, providing low stimulation thresholds, high spatial resolution, and high contrast for pixel sizes down to 20µm. Pillar electrodes offer an alternative for extending past the debris layer. Electroplating of 3D structures is compatible with the fabrication process of flat photovoltaic arrays, enabling much more efficient retinal stimulation.

Three-dimensional electro-neural interfaces electroplated on subretinal prostheses.

Objective: High-resolution retinal prosthetics offer partial restoration of sight to patients blinded by retinal degenerative diseases through electrical stimulation of the remaining neurons. Decreasing the pixel size enables an increase in prosthetic visual acuity, as demonstrated in animal models of retinal degeneration. However, scaling down the size of planar pixels is limited by the reduced penetration depth of the electric field in tissue. We investigate 3-dimensional structures on top of the photovoltaic arrays for enhanced penetration of electric field to permit higher-resolution implants. Approach: We developed 3D COMSOL models of subretinal photovoltaic arrays that accurately quantify the device electrodynamics during stimulation and verified it experimentally through comparison with the standard (flat) photovoltaic arrays. The models were then applied to optimise the design of 3D electrode structures (pillars and honeycombs) to efficiently stimulate the inner retinal neurons. The return electrodes elevated on top of the honeycomb walls surrounding each pixel orient the electric field inside the cavities vertically, aligning it with bipolar cells for optimal stimulation. Alternatively, pillars elevate the active electrode into the inner nuclear layer, improving proximity to the target neurons. Modelling results informed a microfabrication process of electroplating the 3D electrode structures on top of the existing flat subretinal prosthesis. Main results: Simulations demonstrate that despite the conductive sidewalls of the 3D electrodes being exposed to electrolyte, most of the charge flows via the high-capacitance sputtered Iridium Oxide film that caps the top of the 3D structures. The 24 µm height of the electroplated honeycomb structures was optimised for integration with the inner nuclear layer cells in rat retina, while 35 µm height of the pillars was optimized for penetrating the debris layer in human patients. Release from the wafer and implantation of the 3D arrays demonstrated that they are mechanically robust to withstand the associated forces. Histology demonstrated successful integration of the 3D structures with the rat retina in-vivo. Significance: Electroplated 3D honeycomb structures produce a vertically oriented electric field that offers low stimulation threshold, high spatial resolution and high contrast for the retinal implants with pixel sizes down to 20µm in width. Pillar electrodes offer an alternative configuration for extending the stimulation past the debris layers. Electroplating of the 3D structures is compatible with the fabrication process of the flat photovoltaic arrays, thereby enabling much more efficient stimulation than in their original flat configuration.

Cellular migration into a subretinal honeycomb-shaped prosthesis for high-resolution prosthetic vision.

In patients blinded by geographic atrophy, a subretinal photovoltaic implant with 100 µm pixels provided visual acuity closely matching the pixel pitch. However, such flat bipolar pixels cannot be scaled below 75 µm, limiting the attainable visual acuity. This limitation can be overcome by shaping the electric field with 3-dimensional (3-D) electrodes. In particular, elevating the return electrode on top of the honeycomb-shaped vertical walls surrounding each pixel extends the electric field vertically and decouples its penetration into tissue from the pixel width. This approach relies on migration of the retinal cells into the honeycomb wells. Here, we demonstrate that majority of the inner retinal neurons migrate into the 25 µm deep wells, leaving the third-order neurons, such as amacrine and ganglion cells, outside. This enables selective stimulation of the second-order neurons inside the wells, thus preserving the intraretinal signal processing in prosthetic vision. Comparable glial response to that with flat implants suggests that migration and separation of the retinal cells by the walls does not cause additional stress. Furthermore, retinal migration into the honeycombs does not negatively affect its electrical excitability, while grating acuity matches the pixel pitch down to 40 µm and reaches the 27 µm limit of natural resolution in rats with 20 µm pixels. These findings pave the way for 3-D subretinal prostheses with pixel sizes of cellular dimensions.

3D electronic implants in subretinal space: long-term follow-up in rodents.

Photovoltaic subretinal prosthesis (PRIMA) enables restoration of sight via electrical stimulation of the interneurons in degenerated retina, with resolution limited by the 100 µm pixel size. Since decreasing the pixel size below 75 µm in the current bipolar geometry is impossible, we explore the possibility of using smaller pixels based on a novel 3-dimensional honeycomb-shaped design. We assessed the long-term biocompatibility and stability of these arrays in rats by investigating the anatomical integration of the retina with flat and 3D implants and response to electrical stimulation over lifetime - up to 9 months post-implantation in aged rats. With both flat and 3D implants, VEP amplitude decreased after the day of implantation by more than 3-fold, and gradually recovered over about 3 months. With 25 µm high honeycomb walls, the majority of bipolar cells migrate into the wells, while amacrine and ganglion cells remain above the cavities, which is essential for selective network-mediated stimulation of the second-order neurons. Retinal thickness and full-field stimulation threshold with 40 µm-wide honeycomb pixels were comparable to those with planar devices - 0.05 mW/mm2 with 10ms pulses. However, fewer cells from the inner nuclear layer migrated into the 20 µm-wide wells, and stimulation threshold increased over 5 months, before stabilizing at about 0.08 mW/mm2. Such threshold is significantly lower than 1.8 mW/mm2 with a previous design of flat bipolar pixels, confirming the promise of the 3D honeycomb-based approach to high resolution subretinal prosthesis.

Electronic Retinal Prostheses.

Retinal prostheses are a promising means for restoring sight to patients blinded by photoreceptor atrophy. They introduce visual information by electrical stimulation of the surviving inner retinal neurons. Subretinal implants target the graded-response secondary neurons, primarily the bipolar cells, which then transfer the information to the ganglion cells via the retinal neural network. Therefore, many features of natural retinal signal processing can be preserved in this approach if the inner retinal network is retained. Epiretinal implants stimulate primarily the ganglion cells, and hence should encode the visual information in spiking patterns, which, ideally, should match the target cell types. Currently, subretinal arrays are being developed primarily for restoration of central vision in patients impaired by age-related macular degeneration (AMD), while epiretinal implants-for patients blinded by retinitis pigmentosa, where the inner retina is less preserved. This review describes the concepts and technologies, preclinical characterization of prosthetic vision and clinical outcomes, and provides a glimpse into future developments.

Interferometric thermometry of ocular tissues for retinal laser therapy.

Controlling the tissue temperature rise during retinal laser therapy is highly desirable for predictable and reproducible outcomes of the procedure, especially with non-damaging settings. In this work, we demonstrate a method for determining the optical absorption, the thermal conductivity, and the thermal expansion coefficients of RPE and choroid using phase-resolved optical coherence tomography (pOCT). These parameters are extracted from the measured changes in the optical path length (ΔOPL) using an axisymmetric thermo-mechanical model. This allows the calculation of the temperature rise during hyperthermia, which was further validated by imaging the temperature-sensitive fluorescence at the same location. We demonstrate that, with a temperature uncertainty of ±0.9°C and a peak heating of about 17°C following a laser pulse of 20 ms, this methodology is expected to be safe and sufficiently precise for calibration of the non-damaging retinal laser therapy. The method is directly translatable to in-vivo studies, where we expect a similar precision.

Electronic photoreceptors enable prosthetic visual acuity matching the natural resolution in rats.

Localized stimulation of the inner retinal neurons for high-acuity prosthetic vision requires small pixels and minimal crosstalk from the neighboring electrodes. Local return electrodes within each pixel limit the crosstalk, but they over-constrain the electric field, thus precluding the efficient stimulation with subretinal pixels smaller than 55 µm. Here we demonstrate a high-resolution prosthetic vision based on a novel design of a photovoltaic array, where field confinement is achieved dynamically, leveraging the adjustable conductivity of the diodes under forward bias to turn the designated pixels into transient returns. We validated the computational modeling of the field confinement in such an optically-controlled circuit by in-vitro and in-vivo measurements. Most importantly, using this strategy, we demonstrated that the grating acuity with 40 µm pixels matches the pixel pitch, while with 20 µm pixels, it reaches the 28 µm limit of the natural visual resolution in rats. This method enables customized field shaping based on individual retinal thickness and distance from the implant, paving the way to higher acuity of prosthetic vision in atrophic macular degeneration.

Photovoltaic implant simulator reveals resolution limits in subretinal prosthesis.

Objective.PRIMA, the photovoltaic subretinal prosthesis, restores central vision in patients blinded by atrophic age-related macular degeneration (AMD), with a resolution closely matching the 100µm pixel size of the implant. Improvement in resolution requires smaller pixels, but the resultant electric field may not provide sufficient stimulation strength in the inner nuclear layer (INL) or may lead to excessive crosstalk between neighboring electrodes, resulting in low contrast stimulation patterns. We study the approaches to electric field shaping in the retina for prosthetic vision with higher resolution and improved contrast.Approach.We present a new computational framework, Retinal Prosthesis Simulator (RPSim), that efficiently computes the electric field in the retina generated by a photovoltaic implant with thousands of electrodes. Leveraging the PRIMA clinical results as a benchmark, we use RPSim to predict the stimulus strength and contrast of the electric field in the retina with various pixel designs and stimulation patterns.Main results.We demonstrate that by utilizing monopolar pixels as both anodes and cathodes to suppress crosstalk, most patients may achieve resolution no worse than 48µm. Closer proximity between the electrodes and the INL, achieved with pillar electrodes, enhances the stimulus strength and contrast and may enable 24µm resolution with 20µm pixels, at least in some patients.Significance.A resolution of 24µm on the retina corresponds to a visual acuity of 20/100, which is over 4 times higher than the current best prosthetic acuity of 20/438, promising a significant improvement of central vision for many AMD patients.

Pixel size limit of the PRIMA implants: from humans to rodents and back.

Objective.Retinal prostheses aim at restoring sight in patients with retinal degeneration by electrically stimulating the inner retinal neurons. Clinical trials with patients blinded by atrophic age-related macular degeneration using the PRIMA subretinal implant, a 2 × 2 mm array of 100µm-wide photovoltaic pixels, have demonstrated a prosthetic visual acuity closely matching the pixel size. Further improvement in resolution requires smaller pixels, which, with the current bipolar design, necessitates more intense stimulation.Approach.We examine the lower limit of the pixel size for PRIMA implants by modeling the electric field, leveraging the clinical benchmarks, and using animal data to assess the stimulation strength and contrast of various patterns. Visually evoked potentials measured in Royal College of Surgeons rats with photovoltaic implants composed of 100µm and 75µm pixels were compared to clinical thresholds with 100µm pixels. Electrical stimulation model calibrated by the clinical and rodent data was used to predict the performance of the implant with smaller pixels.Main results.PRIMA implants with 75µm bipolar pixels under the maximum safe near-infrared (880 nm) illumination of 8 mW mm-2with 30% duty cycle (10 ms pulses at 30 Hz) should provide a similar perceptual brightness as with 100µm pixels under 3 mW mm-2irradiance, used in the current clinical trials. Contrast of the Landolt C pattern scaled down to 75µm pixels is also similar under such illumination to that with 100µm pixels, increasing the maximum acuity from 20/420 to 20/315.Significance.Computational modeling defines the minimum pixel size of the PRIMA implants as 75µm. Increasing the implant width from 2 to 3 mm and reducing the pixel size from 100 to 75µm will nearly quadrupole the number of pixels, which should be very beneficial for patients. Smaller pixels of the same bipolar flat geometry would require excessively intense illumination, and therefore a different pixel design should be considered for further improvement in resolution.

Interferometric imaging of thermal expansion for temperature control in retinal laser therapy.

Precise control of the temperature rise is a prerequisite for proper photothermal therapy. In retinal laser therapy, the heat deposition is primarily governed by the melanin concentration, which can significantly vary across the retina and from patient to patient. In this work, we present a method for determining the optical and thermal properties of layered materials, directly applicable to the retina, using low-energy laser heating and phase-resolved optical coherence tomography (pOCT). The method is demonstrated on a polymer-based tissue phantom heated with a laser pulse focused onto an absorbing layer buried below the phantom's surface. Using a line-scan spectral-domain pOCT, optical path length changes induced by the thermal expansion were extracted from sequential B-scans. The material properties were then determined by matching the optical path length changes to a thermo-mechanical model developed for fast computation. This method determined the absorption coefficient with a precision of 2.5% and the temperature rise with a precision of about 0.2°C from a single laser exposure, while the peak did not exceed 8°C during 1 ms pulse, which is well within the tissue safety range and significantly more precise than other methods.

A Smartphone-Based Near-Vision Testing System: Design, Accuracy, and Reproducibility Compared With Standard Clinical Measures.

BACKGROUND AND OBJECTIVE: Ophthalmologic telemedicine has emerged during the COVID-19 pandemic. The objective of this study is to assess the accuracy and reproducibility of a smartphone-based home vision monitoring system (Sightbook) and to compare it with existing clinical standards. PATIENTS AND METHODS: Near Snellen visual acuity (VA) was measured with Sightbook and compared with conventional measurements for distance and near VA at an academic medical center ophthalmology clinic in 200 patients with a variety of different specified preexisting ocular conditions. Measurements of contrast sensitivity were also compared by using an existing commercially available chart system in 15 normal patients and 15 patients with age-related macular degeneration. RESULTS: Sightbook VA tests were reproducible (SD = ±0.054 logMAR), and correlation with standard VA methods was significant (R > 0.87 and P < .001). Sightbook contrast sensitivity measurements were reproducible (SD/mean ratio, 0.02 to 0.04), yielding results similar to those of standard tests (R2 > 0.87 and P < .001). CONCLUSIONS: Smartphone-based VA and contrast sensitivity are highly correlated with standard charts and may be useful in augmenting limited inoffice care. [Ophthalmic Surg Lasers Imaging Retina. 2022;53:79-84.].

Vertical-junction photodiodes for smaller pixels in retinal prostheses.

Objective.To restore central vision in patients with atrophic age-related macular degeneration, we replace the lost photoreceptors with photovoltaic pixels, which convert light into current and stimulate the secondary retinal neurons. Clinical trials demonstrated prosthetic acuity closely matching the sampling limit of the 100µm pixels, and hence smaller pixels are required for improving visual acuity. However, with smaller flat bipolar pixels, the electric field penetration depth and the photodiode responsivity significantly decrease, making the device inefficient. Smaller pixels may be enabled by (a) increasing the diode responsivity using vertical p-n junctions and (b) directing the electric field in tissue vertically. Here, we demonstrate such novel photodiodes and test the retinal stimulation in a vertical electric field.Approach.Arrays of silicon photodiodes of 55, 40, 30, and 20µm in width, with vertical p-n junctions, were fabricated. The electric field in the retina was directed vertically using a common return electrode at the edge of the device. Optical and electronic performance of the diodes was characterizedin-vitro, and retinal stimulation threshold measured by recording the visually evoked potentials in rats with retinal degeneration.Main results.The photodiodes exhibited sufficiently low dark current (<10 pA) and responsivity at 880 nm wavelength as high as 0.51 A W-1, with 85% internal quantum efficiency, independent of pixel size. Field mapping in saline demonstrated uniformity of the pixel performance in the array. The full-field stimulation threshold was as low as 0.057±0.029mW mm-2with 10 ms pulses, independent of pixel size.Significance.Photodiodes with vertical p-n junctions demonstrated excellent charge collection efficiency independent of pixel size, down to 20µm. Vertically oriented electric field provides a stimulation threshold that is independent of pixel size. These results are the first steps in validation of scaling down the photovoltaic pixels for subretinal stimulation.

FUNCTIONAL AND STRUCTURAL EFFECTS OF NONDAMAGING RETINAL LASER THERAPY FOR MACULAR TELANGIECTASIA TYPE 2: A Randomized Sham-Controlled Clinical Trial.

PURPOSE: Macular telangiectasia (MacTel) Type 2 is a progressing neurovascular disease of the macula, currently lacking effective treatment. This study assessed the effect of nondamaging retinal laser therapy (NRT) compared with sham. METHODS: Twelve MacTel patients were enrolled in this double-masked, controlled, randomized clinical trial. For the nine patients with both eyes eligible, one eye was randomized to NRT or sham and the other received alternate treatment. For three patients with only one eye eligible, that eye was randomly assigned either NRT or sham. Ellipsoid zone disruption, best-corrected visual acuity, and macular automated perimetry at 12 months served as structural and functional measures. RESULTS: Eleven eyes were randomized to sham and 10 to NRT. Baseline best-corrected visual acuity was 66 letters (20/50) for sham and 72 letters (20/40) for NRT (P = 0.245). Ellipsoid zone disruption area was 298 µm2 in sham and 368 µm2 in NRT (P = 0.391). At 12 months, ellipsoid zone disruption increased by 24% in sham and decreased by 34% in NRT (P < 0.001). Best-corrected visual acuity measures remained stable during follow-up compared with baseline. At 1 year, the mean macular sensitivity was 28 dB in the NRT group, compared with 26 dB in sham. CONCLUSION: Nondamaging retinal laser therapy was safe and well tolerated in patients with MacTel and resulted in structural and functional improvements, which could represent a protective effect of laser-induced hyperthermia. Longer follow-up and larger number of patients should help corroborate these effects.

Decoding network-mediated retinal response to electrical stimulation: implications for fidelity of prosthetic vision.

Objective. Patients with photovoltaic subretinal implant PRIMA demonstrated letter acuity ∼0.1 logMAR worse than sampling limit for 100µm pixels (1.3 logMAR) and performed slower than healthy subjects tested with equivalently pixelated images. To explore the underlying differences between natural and prosthetic vision, we compare the fidelity of retinal response to visual and subretinal electrical stimulation through single-cell modeling and ensemble decoding.Approach. Responses of retinal ganglion cells (RGCs) to optical or electrical white noise stimulation in healthy and degenerate rat retinas were recorded via multi-electrode array. Each RGC was fit with linear-nonlinear and convolutional neural network models. To characterize RGC noise, we compared statistics of spike-triggered averages (STAs) in RGCs responding to electrical or visual stimulation of healthy and degenerate retinas. At the population level, we constructed a linear decoder to determine the accuracy of the ensemble of RGCs onN-way discrimination tasks.Main results. Although computational models can match natural visual responses well (correlation ∼0.6), they fit significantly worse to spike timings elicited by electrical stimulation of the healthy retina (correlation ∼0.15). In the degenerate retina, response to electrical stimulation is equally bad. The signal-to-noise ratio of electrical STAs in degenerate retinas matched that of the natural responses when 78 ± 6.5% of the spikes were replaced with random timing. However, the noise in RGC responses contributed minimally to errors in ensemble decoding. The determining factor in accuracy of decoding was the number of responding cells. To compensate for fewer responding cells under electrical stimulation than in natural vision, more presentations of the same stimulus are required to deliver sufficient information for image decoding.Significance. Slower-than-natural pattern identification by patients with the PRIMA implant may be explained by the lower number of electrically activated cells than in natural vision, which is compensated by a larger number of the stimulus presentations.