Modeling brain functional dynamics via hidden Markov models

Abstract

Functional connectivities constructed via resting state fMRI (R-fMRI) data have been widely used to study the brain's functional activities and to characterize the brain's states. However, the temporal dynamic transition patterns of the brain's functional states have been rarely investigated before. In this paper, we present a novel algorithmic framework to cluster and label the brain's functional states, and learn their hidden Markov models (HMMs). Here, the brain's functional state is compactly represented by a large-scale functional connectivity matrix, called functional connectome state (FCS), and the temporal FCS sequences are derived via an overlapping sliding time window approach. The best-matched HMM learned for ADHD patients revealed a meaningful phenomenon of psychiatric conditions, that is, the tendency to enter into, and inability to disengage from, a negative mood state. Experimental results demonstrated 87% of ADHD patients and 89% of normal controls are successfully classified via multiple HMMs by using majority voting.