We're preparing a study that will require transformation to Talairach coordinates to facilitate comparisons among multiple subjects. My question is, what are the best parameters to use for the image from which the transform will be made? Given that we use axial views, is it necessary to use the coronal and sagittal views generated by AFNI to plan the transform? or is there a way to use independently-acquired coronal and sagittal views? We've been using T1-weighted 256x256x21 images for our anatomicals. Should we try to make the voxel size isotropic by using a larger number of slices? I'm asking these because I've tried to transform some of the data that we already have, and I found that I usually got a good transform of the high-resolution image, but when the transform was automatically applied to the EPI, the results were good only for comic relief. Now we all know, of course, that EPIs usually have some distortion, but this was outrageous. Is there anything that can or should be done to the EPIs to get better results?
