Estimating Large Scale Dynamic Brain Networks with Resting State data from the Human Connectome Project

Sanmi Koyejo, University of Illinois at Urbana-Champaign

Usage Details

Neuroscience seeks to understand how our brains function, and as the human impact of neurological disorders continues to grow, this knowledge holds the potential to improve diagnosis, prognosis and treatment of neurological disorders such as Alzheimer's disease, schizophrenia, traumatic brain injury, among others. Recent scientific progress has shown that the multi-scale interactions between brain regions, typically represented as networks, are a key to understanding biological systems and potential mechanisms that explain diseases and disorders. This proposal aims to address key challenges in data-driven discovery by proposing improved methods for brain network estimation. Our current prototype is able to estimate time varying connectivity between tens of brain regions and hundreds of time points. Blue Waters will enable our methods to scale to thousands of brain regions, and incorporate joint estimations across thousands of subjects leading to meaningful scientific discoveries.