> I've done realignment(estimate and reslice) and coreg(estimate). I've
> used just default setting.
> But I get problem with segmentation.
> The c1 and c2 result of the segmentation are not aligned with structural
Is there any chance that you performed the segmentation, and then
coregistered the structural image? If this were the case, then the
structural image would be moved (during coregistration, in order to
match your functional images) but the segmented images (c1* and c2*)
would not, so they wouldn't line up. What I would do is:
1) Delete the segmented images
2) Use CheckReg to make sure that your structural image is registered
with your functional images (if not, run coregister:estimate)
3) Segment the structural image again
Now the segmented images should be aligned with the structural image
(and the functional images).
> Using Check Reg, I've found tissue probability map images (in "tpm" folder
> of SPM) and structural image are not a good match.
> In that case, do I need to change probability map, or reorient structural
The tissue probability maps are in standard (MNI) space, and so they
will not line up with any individual subject (since any given subject
will not be in MNI space). So, this mismatch is fine. If the
segmentation fails, it can sometimes help to move the subject's
structural image so that it is generally aligned with the template
(origin within ~ 3cm, no extreme rotations). (There are previous
posts on how to do this.) However, if the segmentation looks fairly
reasonable, this isn't really necessary.
(Note that for an fMRI study, if you reorient the structural image to
match the MNI-space template, you'll also need to reorient all of the
functional images, so that for each subject their structural and
functional images are still aligned with each other.)
Hope this helps!
Dr. Jonathan Peelle
Department of Neurology
University of Pennsylvania
3 West Gates
3400 Spruce Street
Philadelphia, PA 19104