Faculty Advisor

Ryder, Elizabeth F.

Faculty Advisor

Shonat, Ross D.

Abstract

A primary task of the retina is to frequency encode light absorption into action potentials. Using simulated artificial neural networks (SANN), the direction selective component of these electrical signals was replicated. The SANN was constructed to mimic the retinal architecture and was trained using published extracellular recordings from ganglion cells. Analysis of the trained model supported the hypothesis that Direction Selectivity arises from asymmetric connections among retinal cells. This project contributes to development of a biology-inspired artificial vision solution.

Publisher

Worcester Polytechnic Institute

Date Accepted

January 2003

Major

Biology and Biotechnology

Major

Biomedical Engineering

Project Type

Major Qualifying Project

Accessibility

Restricted-WPI community only

Advisor Department

Biology and Biotechnology

Advisor Department

Biomedical Engineering

Share

COinS