My best guess is that it is a file format issue to do with the (lack of) image
positioning information in the headers. Use the Display button to position
the images so that the AC is close to the 0,0,0 coordinate (in mm) and the
images are axial (search the archives for more info). This is needed because
the segmentation begins with an affine image registration, and if this fails
then the whole thing goes wrong. Usually, when data are converted from
DICOM, much of this information would be automatically incorporated in the
headers. Similarly, NIfTI format images should generally have this
information in their headers.
Best regards,
-John
Friday 22 May 2009 21:07, Maged Goubran wrote:
> Hi SPMers,
>
> have a quick question:
>
> I've tried segmenting the grey matter of the new 20 BrainWeb anatomical
> models for normal brains on SPM5 but I got black images. And when i changed
> the coordinates of the BrainWeb images I got distoreted (incomplete) grey
> matter !!!!
>
> How can I solve this problem....?
> thanks
|