NEDICA: Detection of group functional networks in FMRI using spatial independent component analysis

Functional magnetic resonance imaging (fMRI) has recently proved its utility in studying brain large-scale networks through fluctuations in resting-state data. To process such rest acquisitions, exploratory methods such as independent component analysis (ICA) are of particular interest. Yet, while successfully applied at the individual level, existing ICA methods still fail to provide robust functional network detection at the group level. In this paper, we propose a method for detecting group functional large-scale networks in fMRI using ICA, which allows to systematically control the consistency of the group results with the individual ones. This approach, called NEDICA (network detection using ICA), was applied on resting-state data from twenty healthy subjects and the robustness of the resulting networks was assessed by a bootstrap sampling procedure. We found seven functional networks that were very representative of the population and highly reproducible on the basis of bootstrap tests. These results were in good agreement with the existing literature and confirmed the ability of fMRI to noninvasively reveal large-scale interactions in the brain.