Hierarchical and graphical analysis of fMRI network connectivity in healthy and schizophrenic groups

Understanding the changes or disruption in the connectivity among brain networks is important for identifying potential biological markers for neuropsychiatric diseases. Multivariate and data-driven methods, especially independent component analysis (ICA), have proven to be a powerful tool in this field. Here, we introduce a novel analysis scheme that incorporates hierarchical and graphical techniques to study the connectivity differences between healthy controls and schizophrenia patients, using the spatial dependence among ICA components as an index of network connectivity. We find that compared to healthy controls, the schizophrenic group presents an altered hierarchy with a number of unusual connections and a significantly decreased small-world index, suggesting disease-related changes in the organization of brain connectivity.