Hi Heidi,

why do they have to be in a common space? You aren't comparing across subjects, just within subject across method. Or are you trying to make maps of accuracy? That will be hard to do unless you use our coordinate system to do your segmentation, which I assume you do not. In any case, mri_extract_label will take a transform such as the talairach.m3z, which will map a label into our atlas coordinates.

cheers Bruce

On Tue, 3 May 2011, Jacobs H (NP) wrote:

Dear FreeSurfers,

We are comparing two methods to measure the volume of a subcortical structure (our method and FreeSurfer). We want to compare them spatially, by doing a voxel-based analyses. We thought to isolate the structure from the aseg.mgh as well as our own structure. Then we smoothed both data sets. Now we are trying to convert the structure from our method to FreeSurfer space (256 x 256). Then, our plan was to put all the data of the structures of both methods all the subjects together (4D) and to run mri_glmfit (comparing method one with method two). Does this seems logically?

But, we were wondering how we should deal with the registration. Because this is in volume space, we cannot register it to the sphere. Registration to the fsaverage may not be very accuracte. Do you have any ideas how to solve this issue?

Thank you very much. Best

Heidi Jacobs

Heidi Jacobs, MSc. PhD student Faculty of Health, Medicine and Life Sciences School for Mental Health and Neurosciences Alzheimer Center Limburg Division Cognitive Neuropsychiatry and Clinical Neurosciences h.jacobs@maastrichtuniversity.nl mailto:h.jacobs@maastrichtuniversity.nl www.maastrichtuniversity.nl http://www.maastrichtuniversity.nl/

Dr. Tanslaan 12, 6229 ET Maastricht P.O. Box 616, 6200 MD Maastricht, The Netherlands T +31 43 38 84 126 F +31 43 38 84 092

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer