Characterization of groups using composite kernels and multi-source fMRI analysis data: Application to schizophrenia

► Complementary sources (GLM, ICA) are combined to better characterize schizophrenia. ► RCK has a lower computing load than other recursive feature elimination algorithms. ► RCK provides a general setting by analyzing nonlinear relationships between voxels. ► Brain regions of segmented whole‐brain data are analyzed and ranked multivariately. ► RCK finds the set of most discriminative brain areas for group classification.