Hi Greg, mri_vol2surf will take a --regheader option which tells it to assume that the movable is in the anatomical space.

doug

On 4/1/11 4:11 PM, GREGORY R KIRK wrote:

Hi guys,

i am using the freesurfer-fsl integration to use labels from the freesurfer cortical surface as seeds for probtrackx probabilistic diffusion tractography. when the output of probtrackx is generated it is already in the freesurfer conformed space. I want to resample the output of probtrackx onto the cortical surface. I have the transformations from the diffusion data to freesurfer conformed space generated using bbregister to do the tractography but to use mri_vol2surf i need to first tranfrorm the output of probtrackx back into the epi space with the inverse of the bbregister tranform ( output in the fsl format) and resample it to the lower resolution epi resolution.

i.e. there is not an option for mri_vol2surf to assume the data to be resampled onto the cortical surface is already in freesurfer conformed space. i suppose i could compute transformations from freesurfer conformed space to say brain.mgz, which would essentially be the identity transformation.

mriconvert -it mgz -ot nii brain.mgz brain.nii i.e. bbregister --mov brain.nii --t1 -s subject -init-fsl --reg register.dat

or i could make a fake identity transform like

1.0 1.0 ? don't know what this number is 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 round ? what should i use here.

whats the best way to go here

thanks

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