Bayesian Modeling of Effective and Functional Brain Connectivity using Hierarchical Vector Autoregressions

TL;DR

This paper introduces a Bayesian hierarchical VAR model for analyzing both functional and effective brain connectivity in resting-state fMRI data, enabling detailed group and individual inferences, especially in ASD research.

Contribution

It presents a novel Bayesian hierarchical VAR approach that models functional and effective connectivity simultaneously, with efficient posterior sampling and group-specific covariance modeling.

Findings

Simplified covariance models overestimate functional connectivity.

Heterogeneous covariance modeling reduces apparent differences in effective connectivity.

The approach effectively distinguishes connectivity patterns in ASD and control groups.

Abstract

Analysis of brain connectivity is important for understanding how information is processed by the brain. We propose a novel Bayesian vector autoregression (VAR) hierarchical model for analyzing brain connectivity in a resting-state fMRI data set with autism spectrum disorder (ASD) patients and healthy controls. Our approach models functional and effective connectivity simultaneously, which is new in the VAR literature for brain connectivity, and allows for both group- and single-subject inference as well as group comparisons. We combine analytical marginalization with Hamiltonian Monte Carlo (HMC) to obtain highly efficient posterior sampling. The results from more simplified covariance settings are, in general, overly optimistic about functional connectivity between regions compared to our results. In addition, our modeling of heterogeneous subject-specific covariance matrices is shown to give smaller differences in effective connectivity compared to models with a common covariance matrix to all subjects.