A0982
Title: Modeling abnormal brain dynamics using statistical physics and MRI
Authors: Alex Leow - University of Illinois (United States) [presenting]
Abstract: While statistical physics has been successfully used to model a wide range of complex systems and phenomena in nature, it has been underutilized in the field of computational neuroimaging. The human brain is governed by fundamental principles of functional segregation and integration; with intrinsic and induced integration directed by a balance of excitatory and inhibitory neural activities. Through the lens of statistical physics, we show how to leverage the mixed-spin ferromagnetic/antiferromagnetic Ising model to reveal patterns of excitation-inhibition (E/I) balance in brain dynamics using multi-modal magnetic resonance imaging (MRI) data (diffusion-weighted MRI and resting-state functional MRI). Consistent with findings from mouse models of Alzheimer's disease (AD) coming out of our own lab and many others, we demonstrated abnormal E/I balance towards hyper-excitation in middle-aged cognitively-normal subjects carrying the Apolipoprotein (APOE-4) allele (a genetic risk for AD) that occurs in the hippocampus.
