Testing different ICA algorithms and connectivity analyses on MS patients

Multiple sclerosis (MS) is a progressive neurological disorder that affects the central nervous system. Functional magnetic resonance imaging (fMRI) has been employed to track the course and disease progression in patients with MS. The two main aims of this study were to apply in a data-driven approach the independent component analysis (ICA) in the spatial domain to depict the active sources and to look at the effective connectivity between the identified spatial sources. Several ICA algorithms have been proposed for fMRI data analysis. In this study, we aimed to test two well characterized algorithms, namely, the fast ICA and the complex infomax algorithms, followed by two effective connectivity algorithms, namely, Granger causality (GC) and generalized partial directed coherence (GPDC), to illustrate the connections between the spatial sources in patients with MS. The results obtained from the ICA analyses showed the involvement of the default mode network sources. The connectivity analyses depicted significant changes between the two applied algorithms. The significance of this study was to demonstrate the robustness of the analyzed algorithms in patients with MS and to validate them before applying them on larger datasets of patients with MS.