Laura, what is the coefficient of variation? Ie, the stddev/mean doug
On 07/11/2013 08:54 AM, L. Koenders wrote:
Dear FreeSurfer experts,
We have a question regarding the aparc output from the stats files, and hopefulle you can help us. We have 4 independent groups, of about 30 subjects per group, all scanned using the same protocal on the same MRI scanner. We ran the cross sectional FreeSurfer (5.2.0, Linux centos4_x86_64) pipeline, and extracted the volume information per person from the .stat files using Matlab. We transfered the volume information into SPSS, to calculate the Means and Standard Deviations per group.
We are interested in a couple of regions of interest (anterior + posterior cingulate, OFC, parahippocamus and insula).
What strange is though, it that the means and standard deviations of 3 groups are comparable across all regions, but the SDs of 1 group are very (very!) large. This is not a group in which we expect brain atrophy or anything like that. Below you see the SD values of all groups, with SD values up to four times bigger in group 1!
1 2 3 4caudal ACC 618 516 643 589 rostral ACC 1583 578 692 623 PCC 1942 721 616 749 OFC 2632 1345 1844 1652 parahip 1198 346 271 394 insula 1410 1033 1096 1017
The outliers are different subject in every region, and I checked the parcellation output of them, that looks fine. We are wondering what could be the cause of this, do you have any ideas?
Thank you in advance,
Kind regards, Laura ________________________________
AMC Disclaimer : http://www.amc.nl/disclaimer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer