Approved Research

High-Dimensional Directed Network Analysis of Human Brains

Lay summary

The human brain is a complex system consisting of many regions that are physically connected through white matter tracts (called structural connectivity) and exert influences on each other to perform various functions. The aims of this research are to (1) reveal interaction patterns, i.e., networks, between all the brain regions for performing different brain functions and (2) investigate how interaction network patterns vary across the population of human subjects and give rise to different human behavior, emotions, and feelings and, thus, different health status. We will develop computationally efficient statistical methods to analyze both functional magnetic resonance imaging (fMRI) and diffusion magnetic resonance imaging (dMRI) data of many subjects from UK Biobank. fMRI measures subjects' entire brain activity, and dMRI measures subjects' structural connectivity between all the brain regions. By utilizing information from both imaging modalities, we will produce efficient estimates of the network interaction patterns between all the regions and reveal the relationship between subjects' brain activity and their health.

We expect the project to last for at least 36 months, given that we need to analyze enormous dMRI and fMRI data under resting state and various tasks of many subjects.

The project results will enhance understanding of the brain's physical and functional organization and its variation across the population of human subjects and contribute critical knowledge of the brain system that generates human emotion, behavior, and feelings. This knowledge will facilitate future research in brain science and assist clinical researchers interested in the neuropathology of many severe neurodevelopmental and neurological disorders, such as Alzheimer's disease and Autism, which have been reported to be related to aberrant brain connectivity.