Hi Doug,
Is it possible to use the bbregister parameters to resample a functional image such that it is in register with the anatomical image but retains the epi geometry? I actually don't care so much about the epi-anat registration, but what I would like to be able to do is motion correct and otherwise preprocess my timeserieses within session* and then move different sessions into a common volumetric space for analysis. As I have great faith in bbregister, ideally the transformation would be determined with this method (I suspect it is superior to picking a slice from one of the sessions to use as a motion correction target across all sessions). Additionally, I would prefer not to upsample the source voxels either into 1x1x1 anatomical geometry or to surface vertices. Unfortunately, my affine understanding isn't deep enough for me to figure this out myself.
Does that make sense, and (if so) do you have any ideas?
Thanks! Michael
* Here i use "session" in the sense of "run", which is different I think from FSFast terminology so apologies for any confusion.