Jason Bard / Mathematics / Faculty Mentor: Pedro Maia
Autism Spectrum Disorder (ASD) can be defined by a difficulty in social interaction alongside repetitive behaviors or interests. As of 2022, the condition affects approximately 1 in 36 children. The Autism Brain Imaging Data Exchange (ABIDE) consists of brain scan data from over 1,000 subjects, with nearly half diagnosed with ASD. This collection includes detailed patient phenotypic information and functional MRI scans taken at rest (rs-fMRIs), which showcases variations in blood flow and brain activity. A preliminary volumetric analysis of all defined brain regions according to the Harvard-Oxford cortical and subcortical max-probabilistic atlases was conducted to estimate the relative neuron count within each neural network component of a modular neural network (MoNN), which mimics the structure of the human brain by keeping the relative size of each region similar in proportion to the real brain. The objective is to assess whether this MoNN can replicate the input time series data accurately by modifying the scaling factor of all regions. In future research, the network’s weight graph will be analyzed to identify distinctive properties that could differentiate between ASD patients and non-ASD individuals. It will also involve integrating a correlation matrix to dynamicize the MoNN and mimic interregional connectivity.
Leave a Reply