Hello all,

I have a mostly-T2 weighted anatomical image (as opposed to the traditional high resolution T1-weighted anatomical), along with EPI data at a lower resolution. For the EPIs, the brain is in a slightly different orientation than the anatomical (tilted forward, a little bit of warping going on, too). I need to align this EPI to the anatomical dataset, and it seems like no matter what I do, the script ends up brutalizing the EPI. It could be because there is not a lot of contrast in the EPI data - so CSF is only slightly identifiable. Most of the cost functions seem to shift the brain to roughly the same space, but not apply any scaling, and not match the CSF components in the same z-direction slice. All data sets are 2D.

If I use the -anat and -epi options, does -anat assume the anatomical is T1 weighted? If so, I assume using -dset1 and -dset2 options would suffice - but then how do I specify a sub-BRIK to use?

Also, does anybody have any suggestions for troublesome dataset alignment?

As always, thanks a ton!