> I have reoriented T1 images in SPM2 and the warning: Shears involved
> are occured. My questions are:
>
> How to quantificate the amount of shearing?
P = spm_get(1,'*.img','Select image');
M = spm_get_space(P);
T = spm_imatrix(M):
T(10:12)
> Does the shearing information can be seen in the Dir Cos matrix?
No. This only shows pure rotations.
> If I rotate these images by a very thiny amount (yaw = 0.001)
> the warning disappear. Could this be an opportunity to avoid
> influences from shearing?
The shears are probably small, but they happen if an image is rotated and then
zoomed. A simple model of zooming to give the right voxel size and then
rigidly rotating, is unable to deal with rotating followed by zooming.
Best regards,
-John
|