Dynamics of functional network organization through graph mixture learning

Ilaria Ricchi, Anjali Tarun, Hermina Petric Maretic, Pascal Frossard, Dimitri Van De Ville

Abstract: Highlights•We estimate multiple functional states with the Graph Laplacian Mixture Model (GLMM).•GLMM learns the graph structure of the states by estimating the Laplacian matrices.•Each state is characterized by a graph and a probability that captures its dynamic.•On task fMRI, GLMM reveals the experimental paradigm unknown to the method.•DMN was found to be the most prominent network and therefore used for comparisons.