Inference of Dynamic Graph Changes for Functional Connectome

Abstract

Dynamic functional connectivity is an effective measure to characterize the brain's responses to continuous stimulus. We propose an inferential method to detect the dynamic changes of brain networks based on the time-varying graphical model. Whereas most existing methods focus on testing the existence of change points, the fluctuations in the brain network structure represent more signals in many neuroscience studies. We propose a novel method to conduct hypothesis tests on the changes in the dynamic brain networks. We introduce a bootstrap statistic to approximate the supreme of the high-dimensional empirical processes over edges. Our simulations show that this framework can successfully capture change points with changed connectivity. Finally, we apply our method to a brain imaging dataset under the natural audio-video stimulus and illustrate that our approach is able to detect the temporal changes in brain networks. The functions of the identified regions are consistent with specific emotional annotations, which are closely associated with changes inferred by the framework.Preprint. Under review.