54. COMPARING NETWORK EFFICIENCY OF AN AUTISM MODEL TO A TYPICALLY DEVELOPING MODEL

Department: Bioengineering
Faculty Advisor(s): Gabriel A. Silva

Primary Student
Name: Vivek Kurien George
Email: vgeorge@ucsd.edu
Phone: 717-209-0107
Grad Year: 2018

Abstract
This paper studies network efficiency in an Autism Spectrum Disorder(ASD) computational model(See ref 2). We compare the network efficiency of the ASD model to a Typically Developing(TD) model. Real world networks are constrained by the internal behavior of its constituents(nodes of various types) and the network's topology(edge connections and spatial locations). A network's topology coupled with its constraints gives rise to interesting dynamics within the network. In this paper, an optimally efficient network is defined as the ratio of the refractory period of a downstream node and the propagation time of a signal along an edge between upstream and downstream nodes, while satisfying a necessary set of conditions(see Ref 1). The computational model simulates the behavior of the primary visual cortex(V1). Divisive normalization is used to vary the behavior of the network from the TD domain to ASD domain. Ref 1: Geometric constraints on the dynamics of networks, by Gabriel Silva Ref 2: A computational perspective on autism, by Ari Rosenberg, Jaclyn Sky Patterson, and Dora E. Angelaki

Industry Application Area(s)
Internet, Networking, Systems | Life Sciences/Medical Devices & Instruments

« Back to Posters or Search Results


Contact:   researchexpo@soe.ucsd.edu   (858) 534-6068