Dear SPM experts,
a while ago I noticed a difference between the masks that I saved from AAL atlas directly and from the AAL atlas within WFU Pickatlas.
Someone pointed out to me that the images are flipped in orientation which can be seen from the matrices in the images headers.
So when I reslice the WFU mask to the AAL mask with SPM 8 the resliced images overlap perfectly.
But when I use the two original masks (not the resliced ones) to visualize my VBM output (which has a voxel size of 1.5 mm) within a ROI I observe different results.
This seems to boil down to the problem that reslicing the two masks to a different voxel size without caring for this flip in orientation yields into two different masks.
For example:
AAL mask of ACC contains 2713
WFU mask of ACC (of AAL atlas) contains 2713 voxels
but
resliced AAL mask of ACC contains 6328 voxels
resliced WFU mask of ACC (of AAL atlas) contains 6439 voxels
Is this a know problem?
Now I'm a bit unsure how to handle the WFU masks correctly.
Reslicing the WFU mask in two steps: 1. to AAL mask (taking the fliped orientation into account) and 2. to the VBM voxel size
results into the same mask as the AAL mask that was resliced to VBM voxel size.
Is this the right way to do? Or should I handle the WFU masks in a different way?
I'm greatful for any hint or comment on this.
Best,
Katharina
|