I am dissatisfied with the segmentation results for the basal ganglia as compared to the cortex. If the cortex is segmented well, the posterior part of the putamen and the thalamus are mostly attributed to white matter. Vice-versa, if the basal ganglia are well segmented, there is too much grey matter in the cortex. I am sure I am not the first to notice this. Is there any way (either curreently implemented or in future plans) to solve this problem? For example: one could select a cube around the basal ganglia and then segment that separately from the cortex, and recombine the images later .... Stefano Marenco, M.D. Senior Staff Fellow, NIMH, CBDB Bldg 10 room 4S235 9000 Rockville PIke Bethesda, MD 20892 email: [log in to unmask] tel: (301) 435-8964 fax: (301) 4807795