Hi Jeff,
we're actively working on the issue of stability of these measures. We have used the non-zero brain voxels, but that's really not ICV, it's more total brain volume. You really can't see the sulcal CSF on a T1, so it will frequently get stripped. If you have global atrophy using this measure may remove it, but that's usually not a problem as it's typically the local effects you are looking for.
cheers, Bruce
On Tue, 23 Sep 2008, Jeff Sadino wrote:
Hello surfers,
We are trying to normalize volumes of different regions of the brain to brain size. By brain size, we would be shooting for everything inside the skull - brain, CSF, and dura. After looking at the four different measurements in aseg.stats and comparing what each of these measures include and also how much each of these measures vary between scans on the same subject, none of them really achieve the consistency that we would like. ICV had a variability of 7.4% with 5 subjects each with 3 scans. The others were at about 3%. We are thinking that if we count the non-zero voxels in brain.mgz, this should be the most accurate measure of ICV. It will include a little bit of the skull and maybe some outside the skull, but since the size of the skull itself should be constant from scan to scan, that should not really matter. This measure gave a variability of only 1.51% Does anybody see any potential errors in using this approach to measure ICV, or suggest improvements or verify its validity?
Thank you so much! Jeff Sadino
mri_label_volume -a brain.mgz
should do the trick.
cheers, Bruce On Mon, 22 Sep 2008, Jeff Sadino wrote:
Hello surfers,
I would like to count the number of non-zero voxels in the image brain.mgz. Does anybody have any ideas what the easiest way to do this would be without using matlab? Thank you all!
Jeff Sadino
See how Windows connects the people, information, and fun that are part of your life. http://clk.atdmt.com/MRT/go/msnnkwxp1020093175mrt/direct/01/ _______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer