Did you check the registrations to see if either or both are inaccurate? As you point out, the eTIV is based on the registration an not an actual count of voxels inside the skull. It may be the case that one time point has more or less neck than the other which could skew the registrations since the registration is done with the skull/neck. doug

On 11/10/13 10:38 AM, Salil Soman wrote:

Hi,

My understanding of etiv is that it is derived from the registration of a brain to atlas space and not actually derived from the freesurfer segmentation. I have a set of patients imaged at 2 time points that all have different etiv values between the first and second scan that are statistically significant. As there was no change in their skulls between scans, I am having difficulty understanding how this is possible. Can anyone suggest what I may be doing wrong?

Best wishes,

Salil

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer