actually I would have agreed with Doug that they should be the same. Can you send us a recon-all.log file from a subject for which they are not (and the 2 stats files I guess)? Just to be clear you ran recon-all a single time on this subject without reprocessing anything?

On Wed, 19 Nov 2008, Margaret Sheridan wrote:

Alexa, et al.,

This also happend to our group (same volumes) Doug Greeve seemed to think they should be the same but my understanding is that they are calculated out of different processing streams so they shouldn't be the same. We posted to the list a week ago asking about this but haven't gotten a reply. Any more insight would be deeply welcome.

Thanks, Margaret

On Wed, Nov 19, 2008 at 12:19 PM, Alexa Nardelli alexan@ualberta.ca wrote:

...

Hi, I have performed subcorticol segmentation using freesurfer and I am now interested in looking at the volume of the hippocampus and the amygdala; the volume statistics under aseg.stats and those under wmparc.stats are different. Is this normal and if so, which ones should I be using?

Thanks

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

Hi Margaret,

the mistake is probably on our end. For now I would recommend just using the aseg.stats hippocampal volumes (and others). They probably don't belong in the wmparc.stats in any case.

cheers, Bruce

On Wed, 19 Nov 2008, Margaret Sheridan wrote:

Bruce,

Thanks for responding & being willing to look at this. I'm not sure if we are making the same mistake as Alexa but we did run recon -all and we got different values for wmparc & aseg (here you can see in wmparc.stats the R hippocampus is 3755 voxels and in aseg it is 3882). I've attached the recon-all.log file.

If there is any other information that would be helpful please let me know.

Cheers, Margaret

On Wed, Nov 19, 2008 at 2:26 PM, Bruce Fischl fischl@nmr.mgh.harvard.eduwrote:

...

actually I would have agreed with Doug that they should be the same. Can you send us a recon-all.log file from a subject for which they are not (and the 2 stats files I guess)? Just to be clear you ran recon-all a single time on this subject without reprocessing anything?

On Wed, 19 Nov 2008, Margaret Sheridan wrote:

Alexa, et al.,

...

This also happend to our group (same volumes) Doug Greeve seemed to think they should be the same but my understanding is that they are calculated out of different processing streams so they shouldn't be the same. We posted to the list a week ago asking about this but haven't gotten a reply. Any more insight would be deeply welcome.

Thanks, Margaret

On Wed, Nov 19, 2008 at 12:19 PM, Alexa Nardelli alexan@ualberta.ca wrote:

Hi, I have performed subcorticol segmentation using freesurfer and I am

...

now interested in looking at the volume of the hippocampus and the amygdala; the volume statistics under aseg.stats and those under wmparc.stats are different. Is this normal and if so, which ones should I be using?

Thanks

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

Hi MOe,

You can make a label file of the region of interest from aseg file.

One easy way to do using matlab, m = MRIread('aseg.mgz'); m.vol = (m.vol == ## | m.vol == ##); MRIwrite(m,'roi.mgz');

## get from FreeSurferColorLUT.txt in freesurfer home.

pratap

Hello,

I'm doing a volume analysis and have found activation in my region of interest. I'm looking to present these data, but there are areas not of interest that also display activation, and so I'd like to leave them out of my image if possible as they are distracting. Does anyone have a suggestion on how I could label the area of interest, and then run an analysis such that activation/deactivation is only displayed in the labeled area?

Thanks, Moe

---

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

Hi Moe,

you could draw a label and use it to mask, but usually you want to show the whole activation pattern so people get a better sense of the data.

cheers, Bruce On Mon, 24 Nov 2008, Zeidan, Mohamed A. wrote:

Hello,

I'm doing a volume analysis and have found activation in my region of interest. I'm looking to present these data, but there are areas not of interest that also display activation, and so I'd like to leave them out of my image if possible as they are distracting. Does anyone have a suggestion on how I could label the area of interest, and then run an analysis such that activation/deactivation is only displayed in the labeled area?

Thanks, Moe