Hi Lily,

I believe bbregister should work with T1 and FLAIR using the --t2 contrast (white darker than grey). Works for me!

You could also use mri_mask with input your parcellation (wmparc) and mask your FLAIR ROI.

Then you could run mri_binarize with "--count file.txt" option for each label with "--match label_id" and it will output the hits on the terminal, but also on your text file. You don't need to save that in a volume, but you would need to do this for each label because if you send many ids on the --match it would merge the count.

Best!

Viv

On Thu, Aug 29, 2019 at 11:18 AM Wang, Lily <LWANG49@partners.org> wrote:

Sorry, I meant the masks were made using FLAIR. Does bbregister work on registering T1 to FLAIR?

I was considering using mri_binarize but because I have 20+ subjects each with multiple ROI's that can span several brain structures, I would have to go through every single subject and manually find the corresponding segmentation code numbers?

Thank you,

Lily

From: freesurfer-bounces@nmr.mgh.harvard.edu <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Greve, Douglas N.,Ph.D. <DGREVE@mgh.harvard.edu>
Sent: Thursday, August 29, 2019 10:55 AM
To: freesurfer@nmr.mgh.harvard.edu <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Corresponding structures to an ROI

Use bbregister instead of mri_robust_register
You can then use mri_label2vol with the --seg option to map into either
direction.
Once they are in register, I would isolate each of your ROIs by creating
a binary volume for each with mri_binarize --i yourrois.mgz --match
SegId --o segid.mgz where SegId is your segmentation code number. Then
run mri_segstats --seg wmparc.mgz --i segid.mgz --ctab-default --sum
sum.segid.dat
The values in the "Mean" field of sum.segid.dat (6th column I think)
will be the fraction of the ROI in that wmparc segment.

On 8/29/19 10:25 AM, Wang, Lily wrote:
> Hi All,
>
> I have several white matter ROI masks unique to each subject that were
> made using the subjects' T2 scans and I would like to use wmparc to
> find out what brain structures those ROI's correspond to. I'm guessing
> that I will need to either register T1 to T2 or T2 to T1 at one point,
> which I have done by using mri_robust_register to register T1 to T2
> (and it seemed to turn out ok, but how do I check other than by eye?
> And is there a better way to register?), and apply that transformation
> to wmparc. But other than figuring out the registration, what else do
> I do?
>
> Thank you,
>
> Lily
>
> _______________________________________________
> Freesurfer mailing list
> Freesurfer@nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

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

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