자료유형  

 학위논문

Control Number  

0017161507

International Standard Book Number  

9798382234540

Dewey Decimal Classification Number  

617.7

Main Entry-Personal Name  

Alex Richard Gogliettino.

Publication, Distribution, etc. (Imprint  

[S.l.] : Stanford University., 2024

Publication, Distribution, etc. (Imprint  

Ann Arbor : ProQuest Dissertations & Theses, 2024

Physical Description  

86 p.

General Note  

Source: Dissertations Abstracts International, Volume: 85-11, Section: B.

General Note  

Advisor: Chichilnisky, E. J.

Dissertation Note  

Thesis (Ph.D.)--Stanford University, 2024.

Summary, Etc.  

요약Interfacing the nervous system with custom electronics that can record and artificially activate neural circuits is a promising way to treat a variety of nervous system disorders. In the retina in particular, epiretinal prostheses are one such neural implant that directly interface with retinal ganglion cells (RGCs) by electrically stimulating them, causing them to send visual information to the brain. These devices are used to treat blindness resulting from photoreceptor degeneration, caused by diseases such as age-related macular degeneration and retinitis pigmentosa. Although modern implants are able to produce light percepts to some degree, the visual percepts are coarse-grained and do not provide useful functional vision, likely because they stimulate RGCs indiscriminately in a manner quite distinct from the natural neural code of the retina. In order to faithfully reproduce the visual neural code of the retina, cellular level control over recording and stimulation are likely required in regions of the retina that support high-acuity vision. In this dissertation, we explore how well visual signals can be reproduced in the central retina of the primate ex vivo using single-electrode epiretinal electrical stimulation as an experimental prototype for a future epiretinal prosthesis.First, we visually stimulate preparations from the central and peripheral retina with a white noise visual stimulus to classify functionally-distinct RGC types, and fit each RGC with a linear-nonlinear Poisson (LNP) model to determine the light response properties of major RGC types in the central retina, and quantitatively compare them to those in the peripheral retina. Next, we electrically stimulate RGCs in the central retina to determine the level of control over RGC responses that can be achieved using single-electrode epiretinal electrical stimulation. Finally, by combining the LNP model parameters and the responses to single-electrode stimulation, we in simulation exploit a recently-developed stimulation algorithm that optimizes stimulation based on visual reconstruction to determine the quality of visual signals that can be reproduced in the central retina, using electrical stimulation. We find that the functional organization, light response and electrical properties of the major RGC types in the central retina are mostly similar to those in the peripheral, with some differences in density, kinetics, linearity, spiking statistics and correlations. The major RGC types could be distinguished by their intrinsic electrical properties. Electrical stimulation targeting parasol cells revealed similar activation thresholds and reduced axon bundle activation in the central retina, but lower stimulation selectivity. Quantitative evaluation of the potential for image reconstruction from electrically evoked parasol cell signals revealed higher overall expected image quality in the central retina. An exploration of inadvertent midget cell activation suggested that it could contribute high spatial frequency noise to the visual signal carried by parasol cells. These results support the possibility of reproducing high-acuity visual signals in the central retina with an epiretinal implant.In the final chapter of this dissertation, we explore how well a recently-developed convolutional neural network model of light responses (Deep Retina) can predict population-level responses of major RGC types in the peripheral primate retina, and provide a quantitative comparison to a linear-nonlinear (LN) model. We find that for models trained and evaluated on naturalistic images, Deep Retina outperformed the LN model in both firing rate predictions as well as the quality of the image features that are represented in the population responses, using an image reconstruction approach.

Subject Added Entry-Topical Term  

Ophthalmology.

Subject Added Entry-Topical Term  

Cellular biology.

Subject Added Entry-Topical Term  

Neurosciences.

Index Term-Uncontrolled  

Linear-nonlinear Poisson

Index Term-Uncontrolled  

Retinal ganglion cells

Index Term-Uncontrolled  

Neural code

Index Term-Uncontrolled  

Central retina

Added Entry-Corporate Name  

Stanford University.

Host Item Entry  

Dissertations Abstracts International. 85-11B.

Electronic Location and Access  

로그인을 한후 보실 수 있는 자료입니다.

Control Number  

joongbu:655455