Independent component analysis of magnetoencephalography data

Independent Component Analysis (ICA) is applied to the Magnetoencephalography (MEG) data of a subject performing a yoga breathing exercise specific for the treatment of obsessive compulsive disorder. The spatio-temporal dynamics observed using a whole-head 148-channel MEG instrument are split into the fundamental modes, thus isolating separate brain activity signals. Experiments were performed on data from different brain regions. Spectral analysis of the more significant signals are presented. Moreover a new tool is developed as a Matlab toolbox to support the scientist both in the visualization and computation phases.