Hello,
The comments in spm_VOI.m in SPM8 (which I understand to be the function
that returns the "small volume" p-values in the Results panel) has the
following comment about using an external mask:
% External mask images should be in the same orientation as the SPM
% (i.e. as the input used in stats estimation). The VOI is defined by
% voxels with values greater than 0.
Could someone please elaborate briefly as to what is meant by the
requirement that the mask "be in the same orientation as the SPM"?
Looking at the code for spm_VOI it appears that the qform matrix of the
mask is accounted for by the following lines (L121-124):
XYZ = D.mat \ [xSPM.XYZmm; ones(1, size(xSPM.XYZmm, 2))];
j = find(spm_sample_vol(D, XYZ(1,:), XYZ(2,:), XYZ(3,:),0) > 0);
XYZ = D.mat \ [ XYZmm; ones(1, size( XYZmm, 2))];
k = find(spm_sample_vol(D, XYZ(1,:), XYZ(2,:), XYZ(3,:),0) > 0);
These lines seem to interpolate the voxels from the mask into the space
of the SPM using nearest neighbor interpolation. So, it seems that the
orientation (i.e., LAS, LPI, RSA, etc), FOV, and voxel size of the SVC
mask should be irrelevant, and that all the necessary matching of the
mask to the space of the SPM is handled automatically within the code of
spm_VOI.m itself. Is that not the case? Or, is the comment above
within spm_VOI.m just left over from some earlier version of the
function?
thanks,
-MH
--
Michael Harms, Ph.D.
--------------------------------------------------------------------
Conte Center for the Neuroscience of Mental Disorders
Washington University School of Medicine
Department of Psychiatry, Box 8134
Renard Hospital, Room 6604 Tel: 314-747-6173
660 South Euclid Ave. Fax: 314-747-2182
St. Louis, MO 63110 Email: [log in to unmask]
--------------------------------------------------------------------
|