Investigative Ophthalmology & Visual Science Cover Image for Volume 61, Issue 7
June 2020
Volume 61, Issue 7
Free
ARVO Annual Meeting Abstract  |   June 2020
Modeling and decoding network-mediated retinal response to electrical stimulation: implications for fidelity of prosthetic vision
Author Affiliations & Notes
  • Elton Ho
    Department of Physics, Stanford University, Stanford, California, United States
    Hansen Experimental Physics Laboratory, Stanford University, Stanford, California, United States
  • Alexander Shmakov
    Computer Science, University of California, Irvine, Irvine, California, United States
  • Daniel V Palanker
    Hansen Experimental Physics Laboratory, Stanford University, Stanford, California, United States
    Department of Ophthalmology, Stanford University, Stanford, California, United States
  • Footnotes
    Commercial Relationships   Elton Ho, None; Alexander Shmakov, None; Daniel Palanker, Pixium Vision (C), Pixium Vision (P), Stanford University (P)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2020, Vol.61, 2204. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Elton Ho, Alexander Shmakov, Daniel V Palanker; Modeling and decoding network-mediated retinal response to electrical stimulation: implications for fidelity of prosthetic vision. Invest. Ophthalmol. Vis. Sci. 2020;61(7):2204.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : Patients with the photovoltaic subretinal implant PRIMA demonstrated letter acuity ~0.1 logMAR worse than the sampling limit for 100μm pixels (1.3 logMAR), while healthy subjects perform by ~0.2 logMAR better than the sampling limit at equivalently pixelated images. To explore the underlying differences between natural and prosthetic vision, we compare the fidelity of the retinal response to visual and subretinal electrical stimulation through single-cell modeling and population decoding.

Methods : Responses of the retinal ganglion cells (RGC) to optical or electrical (1mm diameter arrays, 75μm pixels) white noise stimulation in healthy and degenerate rat retinas were recorded via MEA. We compared statistics of the spike-triggered average (STA) in RGCs responding to electrical or visual stimulation of healthy and degenerate retinas. Each RGC was fit with linear-non-linear (LN) and convolutional neural network (CNN) models. At the population level, we constructed a linear decoder to determine the certainty with which the ensemble of RGCs can support N-way discrimination tasks.

Results : Electrical receptive fields of the RGCs include ON and OFF responses with antagonistic surround, similar size to the natural receptive fields. However, response to electrical stimulation is significantly noisier than natural signaling. The signal-to-noise ratio of electrical STAs in degenerate retinas matched that of the natural responses when 80% of the spikes were replaced with random timing. Although LN and CNN models can fit natural visual responses very well, they fail in predicting spike timings elicited by electrical stimulation in most RGCs. For population response, a decoder trained on electrical data of 19 cells can perform the 4-way Landolt-C orientation test at 50% accuracy, 15% lower than with visual responses of 19 cells in healthy retina, and 25% lower than with 49 cells over the same area.

Conclusions : RGC responses to subretinal electrical stimulation are significantly more stochastic than in natural vision, especially the spike timing. However, adequately large population of cells may provide sufficient information for decoding the applied stimulus, albeit with lower accuracy than natural vision due to smaller number of responsive cells and higher firing stochasticity. These findings shed light on the lower fidelity of prosthetic vision at the same pixilation level.

This is a 2020 ARVO Annual Meeting abstract.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×