Hi Bruce,
Thanks for getting back to me! I am using my own ROIs, created based on an ALE (Activation Likelihood Estimation) conducted in the gingerALE software, and dividing the ALE map into smaller clusters in Marsbar (SPM).
I then applied the ROIs to the surface based registration(fsaverage registration) and turned these into label files (mri_cor2label) and extracted volumes from my participants (mri_label2label)
I noticed that some of my biggest ROIs didn't come out with the larges volumes in my final data and started looking at the correlations of my volumes across the different steps. I found the volumes of my templates in SPM (Marsbar) and after the projection to fsaverage were highly correlated although the ROIs where much smaller in fsavarage. But there was no correlation between my ROIs in fsaverage and my extracted data (based on 500 participants).
I did not correct for the reduced surface area in these analyses, so thanks a lot for making me aware of this. Can you by any chance direct me to a reference for this? And do you think I should expect a correlation between the templates in fsaverage my data if I do this? I am not interested in there correlations as such, but ran these analyses to check my data because I am worried something might have gone wrong.
Thanks a lot!
Best,
Sarah
________________________________________ From: Bruce Fischl fischl@nmr.mgh.harvard.edu Sent: 07 October 2013 13:21 To: Jensen, Sarah Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Consistency between volumes in fsaverage and subject data?
Hi Sarah
what ROIs do you mean? And did you correct for the reduced surface area in fsaverage w.r.t. the individual surfaces? If not, it's not surprising that fsaverage ROIs have less surface area. It is an average surface and less folded than any individual.
cheers Bruce
On Mon, 7 Oct 2013, Jensen, Sarah wrote:
Hi Freesurfer users,
I am analysing some volumetric data and have noticed that the volumes I extract from my participants are very different from the volumes of my templates in fsaverage space. I for instance have two ROIs that stand out as big (about 6-8 times larger that the other ROIS) in fsaverage space, but come out as almost indistinguishable from the other ROIs in my data. When I sort my ROIs according to size there seems to be no consistency between the size in fsaverage space and my subjects, and the correlation between the volume in fsaverage and my subjects is about .05 - 0.1.
I am new to freesurfer, and would intuitively expect greater consistency between the ROIs in fsaverage and the subjects. Has anyone else had similar experiences, and is this lack of consistency between volumes in fsaverage space and participant data perfectly normal?
Thanks,
Sarah
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.