Hi Bruce,
I'm using v.4.4. Is there any processing that I can do to remove the bias without having to rerun everything with the newest version?
Thanks, Liz
-----Original Message----- From: Bruce Fischl [mailto:fischl@nmr.mgh.harvard.edu] Sent: Monday, August 31, 2009 2:17 PM To: Elizabeth Selgrade Cc: 'freesurfer@nmr.mgh.harvard.edu' Subject: Re: [Freesurfer] Longitudinal vs. Cross-sectional Volumes
Hi Liz,
what version are you using? Our newest longitudinal version removes this bias.
cheers, Bruce
On Mon, 31 Aug 2009, Elizabeth Selgrade wrote:
Hello FreeSurfer Users,
I have been using the longitudinal stream to process four timepoints for about 20 subjects. I've extracted 9 ROIs and gotten their volumes for both the longitudinal segmentations and the cross-sectional segmentations. I'm noticing that there are seemingly significant differences in the volumes for the two processing streams. The cross-sectional stream gives results that range from about 0.5% to 19%
above the longitudinal volumes.
Have any of you gotten similar results?
I'm curious as to why this is happening. I'm using different tools to register the volumes to two different spaces - tkregister2 & mri_label2vol move my cross-sectional aseg.mgz's to native space. Mri_convert moves the longitudinal aseg.mgz's to the base space. I don't think the different spaces are the root of the problem, as they
should have the same resolution.
Could the different tools be affecting the size of my ROIs?
Or, could initializing the longitudinal segmentation with the base image somehow alter the volumes?
Thank you! Liz Selgrade
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer