Analysis of time-varying brain connectivity using nonparametric Bayesian model
1 Faculty of Information Technology, Sao Do University, HaiDuong, Vietnam.
2 Faculty of Information Technology, Hanoi University of Natural Resources and Environment, HaNoi, VietNam.
3 Center for Information Technology Application, University of Transport and Communications, HaNoi, VietNam.
Research Article
Global Journal of Engineering and Technology Advances, 2024, 21(03), 069–076.
Article DOI: 10.30574/gjeta.2024.21.3.0228
Publication history:
Received on 28 October 2024; revised on 14 December 2024; accepted on 16 December 2024
Abstract:
The study of time-varying brain connectivity is essential for understanding the dynamic interactions between different brain regions, especially in the context of cognitive processes, neurological disorders, and brain network functioning. In this paper, we present a novel approach for analyzing effective brain connectivity using a nonparametric Bayesian model. Specifically, we apply a Hierarchical Dirichlet Process Auto-regressive Hidden Markov Model (HDP-AR-HMM) to capture the temporal evolution and structural patterns of connectivity between brain regions. The proposed model allows for flexible, data-driven clustering of brain states while incorporating both temporal dependencies and hidden states. We demonstrate the utility of this method in revealing the dynamic structure of brain networks and uncovering time-varying patterns of effective connectivity. Our approach is validated using Alzheimer fMRI data, showing that it the dynamic interaction among brain regions during a simple sensory-motor task experiment, providing new insights into the dynamic processes governing brain activity.
Keywords:
Brain Connectivity; Time-varying Connectivity; Hierarchical Bayesian modeling; fMRI; HDP
Full text article in PDF:
Copyright information:
Copyright © 2024 Author(s) retain the copyright of this article. This article is published under the terms of the Creative Commons Attribution Liscense 4.0