It depends on what the reviewer is concerned about. You don't want to use a fully non-rigid registration because the drop out will look like the edge of the brain, and the non-rigid will warp it incorrectly. If he/she is talking about using something more constrained like B0 correction, then it is possible that some things could change. In general, where you have B0 distortion you also have dropout, so correcting for B0 distortion often has a very small effect (no ref on that unfortunately). doug

On 07/30/2012 11:01 AM, SHAHIN NASR wrote:

Hi,

  In our functional MRI studies, we co-registration the functional 

images to anatomical scans (a part of preproc-sess routine in version 5) before measuring the contrasts. But for one of our studies, a reviewer has asked if a “non-rigid image registration” would change results when we are assessing activity within artifact-prone brain areas (e.g. anterior temporal areas).

  As far as I know, freesurfer registration works fine and even in 

this study, we have found very strong activity/contrast within anterior tempotal lobe. But I was wondering whether you have tested this possibility or not. If yes, then I appreciate if you can introduce me a reference (published article) to support our answer.

Regards

-- Shahin Nasr

PhD in Cognitive Neuroscience Martinos Imaging Center, MGH Harvard Medical School