External Email - Use Caution
Dear Everyone,
A kind reminder. I'm unable to look beyond this step and am really looking for some help. I would like to know if we can use the probabilistic information directly on mri_train
Thank you,
Kind regards
Mageshwar Selvakumar, M.Sc
Department of Pharmakologische Bildgebung und Bildanalyse
Friedrich Alexander Universität Erlangen Nürnberg
Office: Fahrstr. 17, 91054 Erlangen
Phone: + 49-9131-8522898
________________________________ From: Mageshwar Selvakumar mageshwar.selvakumar@outlook.com Sent: Monday, April 23, 2018 10:36 AM To: freesurfer@nmr.mgh.harvard.edu; Bruce Fischl Subject: Query regarding creation of atlas directly from probabilistic information
Dear All and Bruce,
This mail is on top of the query I had a couple of weeks back where I misinterpreted that I had binary mask of each structure as individual nifti volume files but on closer look, the nifti files had the probability that the structure is present at that certain location. On looking into the freesurfer website, I came across this documentation of creation of Atlas, https://surfer.nmr.mgh.harvard.edu/fswiki/SurfaceLabelAtlas SurfaceLabelAtlas - Free Surfer Wikihttps://surfer.nmr.mgh.harvard.edu/fswiki/SurfaceLabelAtlas Overview. This document provides an overview of two related topics: How the FreeSurfer pipeline automatically labels a subject's regions (producing an annotation file) by comparing the subject's curvature data to an "atlas" ("gcs") file. surfer.nmr.mgh.harvard.edu
In this documentation the process is done by looping through N subjects to get the probability information, and these subjects are registered to the registration template derived from average of reference templates. I would like to know if I could use these probability information directly and since the atlas information is already in the MNI152 space, can I use the registration template which is already available with freesurfer, and if yes, can someone guide me on which registration files I must use.
Secondly, I tried using mri_vol2surf and tried to convert the volume data to a surface data before creating labels, but on doing so I had errors when creating labels from the surface files. To convert into volume I ran the below code:
*mri_vol2surf --mov ID076_LL_PHGa_l_HvOx_prob.mgz --ref ID076_LL_PHGa_l_HvOx_prob.mgz --out ID076_LL_PHGa_l_HvOx_prob_surf.mgz --hemi lh --mni152reg --projdist-max -2 0 .1 --reshape --interp nearest*
Then I to convert this surface file into label I ran
*mri_cor2label --i ID080_CPuP_l_HvOx_prob_surf.mgz --l area1.label --id 076*
And I got the error
*Found 0 label voxels
ERROR: found no voxels matching id 76*
But the process runs fine if I run the mri_cor2label on the volume file. It would be nice if someone can help me out on where I'm going wrong here.
Thank you,
Kind regards
Mageshwar Selvakumar, M.Sc
Department of Pharmakologische Bildgebung und Bildanalyse
Friedrich Alexander Universität Erlangen Nürnberg
Office: Fahrstr. 17, 91054 Erlangen
Phone: + 49-9131-8522898