Hi
> On 5 Feb 2019, at 21:41, Matthias Heyne <[log in to unmask]> wrote:
>
> Dear Experts,
>
> We're using probtrackx2 and custom cortical ROIs to run subcortical parcellation of various subcortical ROIs output by Freesurfer's recon-all command. We've noticed that using target ROIs (--targetmasks flag) supplied as volume masks generated at 2mm white matter depth (using Freesurfer's mri_aparc2aseg command) leads to results different from using target ROIs specified as surface labels (Freesurfer .label files converted to .gii format).
> Would anyone be able to explain how probtrackx2 transforms these surface ROIs so that it can estimated seed-to-seed connection probabilities between seed and target voxels? E.g. do the surface vertices get mapped directly into 3D diffusion space using nearest neighbor or some other metric?
Internally, probtrackx2 uses the transformations provided (—xfm and —invxfm) to send things back and forth between seed space and diffusion space. There is no interpolation in this process. The only interpolation is done when deciding to store sample streamlines. as these are stored as visitation maps (histograms) we use nearest neighbour for that.
Cheers
Saad
>
> Here are two sample probtrackx2_gpu commands (preceding the find_the_biggest commands):
>
> using surface-based targets:
> probtrackx2_gpu -s [.../dmri.bedpostX/merged] -x [subcortical volume mask in freesurfer space] - m [.../dmri/nodif_brain_mask.nii.gz] --targetmasks=[textfile of full paths to .gii files] --os2t --xfm=[freesurfer to FA space transposition matrix] --meshspace=freesurfer --seedref=[freesurfer orig.nii.gz volume] --dir=[output directory] --avoid=[combined volume mask of ventricles + CSF in freesurfer space] --savepaths --onewaycondition [lots of flags specifying defaults -> for our record] --forcedir
>
> using volume masks as targets:
> probtrackx2_gpu -s [.../dmri.bedpostX/merged] -x [subcortical volume mask in diffusion space] - m [.../dmri/nodif_brain_mask.nii.gz] --targetmasks=[textfile of full paths to volume mask files in diffusion space] --os2t --dir=[output directory] --avoid=[combined volume mask of ventricles + CSF in diffusion space] --savepaths --onewaycondition [lots of flags specifying defaults -> for our record] --forcedir
>
> Thank you very much!
> Matthias
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|