Thanks for that, Doug and Eugenio. I have two follow up queries:

 

1. I tried your command (with the appropriate paths) but it didn’t work (error below), however, after adding the flag -at it worked. I wanted to make sure this was still warping the hippocampal/amygdala segmentation using the CVS MNI152 registration.

Command you suggested: mri_convert lh.hippoAmygLabels-T1.v21.CA.FSvoxelSpace.mgz final_CVSmorph_tocvs_avg35_inMNI152.m3z output-in-mni152.mgz
Error: mri_convert: extra argument ("output-in-mni152.mgz")

Command with flag (which worked): mri_convert lh.hippoAmygLabels-T1.v21.CA.FSvoxelSpace.mgz -at final_CVSmorph_tocvs_avg35_inMNI152.m3z output-in-mni152.mgz

 

2. Providing this is correct, I then have a query about the transform into MNI152 space. The original amygdala segmentation gives me 9 areas (attached image: original segmentation), with their relevant labels (7001/3/5-10 and 7015); however, the transformed MNI152 amygdala segmentation doesn’t match these labels (attached imaged: segmentation in MNI). Instead:
1. For the left hemisphere I get 9 areas, labelled from 7001-9, I am missing the AAA and paraluminar nucleus, but there are labels for the basolateral and centromedial nuclei.
2. For the right hemisphere I get 10 areas, labelled from 7001-10, I am missing the paraluminar nucleus, and there are still the additional basolateral and centromedial nuclei.

Is this somehow correct or has something gone horribly wrong?

 

Thanks so much for all the help.

Elizabeth

 

 

        External Email - Use Caution        

Hi Eugenio,

 

Just to double check I understand:

 

1. If I run segmentHA_T1.sh on the output from recon-all, then the output from segmentHA_T1.sh is also in fsaverage space (not native space)?

None of the output is in fsaverage space

2.  

1. If this is correct, then can I just warp this segmentation (from fsaverage space to MNI152 space) using mri_cvs_register?

                                                                                       i.      How would I specify that mri_cvs_register should run on the registration given by cvs for each subject (in the flagged arguments it only shows how to specify registration to MNI152 space)?

OR

3. Did you mean that I can specify segmentHA_T1.sh to run in native space (I couldn’t find anything on how to do this) and then warp this native segmentation to MNI152 space?

It always runs in native space. You can use CVS to warp ti to MNI space with
mri_convert lh.hippoAmygLabels-T1.v21.CA.FSvoxelSpace.mgz final_CVSmorph_tocvs_avg35_inMNI152.m3z output-in-mni152.mgz

4.  

 

Many thanks for the help and clarification,

 

Elizabeth

 

From: <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of "Iglesias Gonzalez, Juan E." <JIGLESIASGONZALEZ@mgh.harvard.edu>
Reply to: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Date: Thursday, 28 May 2020 at 10:20 pm
To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Subject: Re: [Freesurfer] Running amygdala segmentation on a warped brain (fsaverage-to-MNI152)

 

Hi Elizabeth,

segment_T1.sh uses a bunch of files in native space for initialization that you’d also have to deform, if you wanted to run it in MNI space.

I think the easiest would be to run it in native space, and then warp the segmentation to MNI, using e.g., the registration given by CVS.

Cheers,

/Eugenio

 

 

Juan Eugenio Iglesias

Senior research fellow

CMIC (UCL), MGH (HMS) and CSAIL (MIT)

http://www.jeiglesias.com 

 

 

 

From: <freesurfer-bounces@nmr.mgh.harvard.edu> on behalf of Elizabeth Haris <e.haris@unsw.edu.au>
Reply-To: Freesurfer support list <freesurfer@nmr.mgh.harvard.edu>
Date: Wednesday, May 27, 2020 at 23:11
To: "freesurfer@nmr.mgh.harvard.edu" <freesurfer@nmr.mgh.harvard.edu>
Subject: [Freesurfer] Running amygdala segmentation on a warped brain (fsaverage-to-MNI152)

 

        External Email - Use Caution        

Hi guys,

 

I’m wanting to run segmentHA_T1.sh on an MNI152 registered brain. On my subject, I have run recon-all, then run mri_cvs_register to obtain a brain image registered in MNI152 space. My questions are:

 

1. In recon-all, is there a way to specify registration to MNI152 space, or is it necessary to first run recon-all and then to run mri_cvs_register?
2. Is it possible to run the segmentHA script on the brain registered in MNI152 space? I’m unable to find where I would specify the use the cvs/norm.mgz (instead of the mri/norm.mgz file). And from what I can see in the script, the parameters set in lines 179/180/181 (below) point to fsaverage brain files, which don’t appear to be the correct targets for the new MNI152 brain; however, there are no corresponding output files in the cvs folder, so I am unsure how to define these parameters.
3. Is there anything I have missed and would also need to do to run the segmentHA script on an MNI152 brain? (I have previously run the script on the average brain and all went well.)

 

179         set ATLASMESH="$FREESURFER_HOME/average/HippoSF/atlas/AtlasMesh.gz";

180         set ATLASDUMP="$FREESURFER_HOME/average/HippoSF/atlas/AtlasDump.mgz";

181         set LUT="$FREESURFER_HOME/average/HippoSF/atlas/compressionLookupTable.txt";

 

Thanks in advance,

 

Elizabeth

_______________________________________________

Freesurfer mailing list

Freesurfer@nmr.mgh.harvard.edu

https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer