Dear Juenger,
Are you referring to the voxel size of in table to be sure that the sizes
are the same? Although both the voxel size and FWHM reported in the table
are the same for both analyses, the *actual* FWHM used to compute the resel
count within the ROI may *not* be the one reported in the table. The voxel
size in the table is for the original map, not for the mask.
Looking at the spm_VOI.m, we see that the fields xSPM.FWHM and xSPM.VOX
don't change after loading the mask, and these are the fields used to
generate the table (spm_list.m, line 344). However, the actual FWHM used to
re-estimate the number of resels is not the same as the xSPM.FWHM when there
are different voxel sizes.
I haven't used the PickAtlas, but I think it produces masks 2x2x2 mm (unless
you resample them later). Indeed, it is when the resolution of the map is
higher than the resolution of the mask that there may be an increase in the
amount of resels, exactly what you are observing!
Hope this can help!
Kind regards,
Anderson
|