Figure 1.

The pipeline of feature extraction and cross-validation Multiple Kernel Learning (MKL) classification. T1/2WI, T1 weighted imaging and T2 weighted imaging; rsfMRI, resting-state fMRI; DTI, diffusion tensor imaging; FA, fractional anisotropy; MD, mean diffusivity; LD, longitudinal diffusivity; TD, transverse diffusivity; RF, random forest; SVM, support vector machine. The used atlases include Desikan–Killiany atlas, fuzzy-cluster parcels, subcortical regions, AtlasTrack, and Gordon parcellations; the morphological features include subcortical region volume, cortical volume, thickness, area, and sulcal depth.