Functional Connectivity in the Resting Brain Revealed by Group-Level Independent Component Analysis
Chen S, Yang Y, Stein E, Ross T, Chuang K
National Institute on Drug Abuse, NIH, National Tsing-Hua University
Coherent low-frequency fluctuations in the resting-state fMRI signal have been used to study functional connectivity. The majority of these studies have utilized hypothesis-driven methods, such as cross-correlation analysis with presumed seed points, to obtain brain connectivity maps. Independent component analysis (ICA) is a data-driven method and has been recently used to reveal resting-state functional networks without a priori seed points. In this study, we used group ICA (gICA) to analyze resting-state fMRI data to improve sensitivity and reduce bias from individual differences. Our results show that patterns of connectivity, including the default mode of brain networks, can be consistently obtained by gICA.