I see this does output a mask.mgh which is just what I need. I can convert this and have my nifti mask. Thanks again.
Doug Greve wrote:
If you're using glmfit, you can --label $SUBJECTS_DIR/fsaverage/lh.Medial_wall.label --mask-inv
Morgan Hough wrote:
Hi Nick,
I see. I am just not sure how to put this back together again to match the registered thickness files in the .mgh file. Do the vertices have unique ids? Can I cat and sort them? Can I dump matching information from my particular thickness group file and then sort?
Is there someway to binarize the .annot file to use as a mask with mri_glmfit?
Cheers,
-Morgan
Nick Schmansky wrote:
Morgan,
If you do this:
mri_annotation2label --subject <subjid> --hemi rh \ --outdir ./labels
you will get label files in the ./labels directory. Those label files are text files listing the vertices and x,y.z of each label, including 'unknown' and 'corpuscallosum' (which are the only two that need excluding). Will that provide you with what you need?
Nick
On Fri, 2007-04-06 at 21:35 +0100, Morgan Hough wrote:
I am using ICA on what would be input to mri_glmfit but since ICA is multivariate I need to exclude regions that don't have interpretable thickness or distortion measures. It seems like unknown and corpus callosum are the two to exclude. Are there any others I should exclude?
I would like to convert the ?h.aparc.annot files to .mgz if that gets me surface files integer coded with the regions. That way I can binarize and exclude those regions with avwtools as I will need to convert to nifti to do the ICA analysis anyway. What would you suggest for this?
Cheers,
-Morgan
Bruce Fischl wrote:
which parcellation are you using?
On Fri, 6 Apr 2007, Morgan Hough wrote:
Thanks Bruce,
Are there any other parcellation units that should be excluded? That seems to be the only white matter PU.
Cheers,
-Morgan
Bruce Fischl wrote:
> Hi Morgan, > > no thickness should not be used in the cc. > > Bruce > On Fri, 6 Apr 2007, Morgan Hough wrote: > > >> I would like to make a mask for group analysis that excludes >> the unknown parcellation area. Since I would like to apply the >> same mask to all subjects (I have already brought them into >> alignment with mri_preproc), I was wondering how I can convert >> the average template label/?h.aparc.annot to a .mgz files that >> I could then convert into a mask. Is there an alternative way >> to produce such a mask? >> >> As a follow-up question, I was wondering if parcellation areas >> such as the corpus callosum should also be excluded from >> analysis? Are the thickness estimates made in the corpus >> callosum parcellation area interpretable? >> >> Thanks in advance for your time. >> >> Cheers, >> >> -Morgan >> _______________________________________________ >> 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