Thank you for your reply.
We did find registration in nonlinear worked better for our sample.
One last question, since I already ran many ROIs and many subjects using FLIRT to transform the ROI to subject's space, would you recommend re-doing it using std2imgcoord? I understand you may not be able to tell; it might be something I need to check and see how it looks, compare using std2imgcoord to what I've done, etc. But as a general question, I suppose you would re-do the ROI generation using std2imgcoord to be consistent? Thanks again!
|