Hi, I have a set of ROIs from SPM in MNI space that I want to load into FSL. When I take the standard FSL brain MNI152_T1_1mm_brain and load it in MRIcro and overlay the ROI, it looks fine, like the first screenshot: http://www.ma-tea.com/~Matt/MRIcroOnMNI.png However, if I resample the 4X4X4mm ROI so it can be overlayed on this brain in FSLView, using: flirt -in 4X4X4mmROI -ref MNI152_T1_1mm_brain -applyxfm -init eye.mat -out 1X1X1mmROI where eye.mat is the identity matrix: 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 The result is incorrect, shown by the second screenshot in FSLView: http://www.ma-tea.com/~Matt/FSLOnMNI.png The ROI file header is the following: <nifti_image nifti_type = 'ANALYZE-7.5' image_offset = '0' ndim = '4' nx = '41' ny = '48' nz = '35' nt = '1' dx = '4' dy = '4' dz = '4' dt = '1' datatype = '16' nbyper = '4' byteorder = 'LSB_FIRST' cal_min = '0' cal_max = '1' descrip = 'FSL4.0' sform_code = '2' sto_xyz_matrix = '-4 0 0 80 0 4 0 -112 0 0 4 -52 0 0 0 1' num_ext = '0' /> I would appreciate any assistance with this as I am unfamiliar with this type of issue if it has to do with the analyze file format or header. Thanks, Matt.