After re-alignment of each subject’s T1- and T2-weighted images, the expectation maximization segmentation method (Van Leemput et al., 1999) was applied to the T1-weighted images to segment the brain into WM, GM, and CSF after co-registration and correction for bias field intensity variation of the T1-weighted images. The major lobes and subcortical regions were then parcellated by overlaying the tissue maps on a reference atlas containing landmarks of 36 brain regions (see (Studholme et al., 2001)), including the regions reported here. The volumes of total cortical GM and total lobar WM were calculated by summing the respective GM and WM values from the major lobes. The intracranial volume (ICV) was estimated by summing all regional tissue and CSF volumes.
