Hi, all.
I want to register CBF maps derived using pCASL to freesurfer's surface and derive CBF values in the same regions where I report my cortical thickness, and volumes. Hence, I did the following: 1) bbregister --s T1_data --mov asldata_mc_even_brain.nii.gz --reg mean_asl_to_fs_T1 --bold --init-fsl
asldata_mc_even_brain is the mean of the control images
2) mri_vol2vol --mov perfusion_calib.nii.gz --targ T1_data/mri/T1.mgz --interp nearest --o perfusion_CBF_to_T1.nii.gz --reg register.dat --no-save-reg
where perfusion_calib is the CBF map in the same space asldata_mc_even_brain.
3) mri_vol2surf --mov perfusion_CBF_to_T1.nii.gz --ref orig.mgz --regheader T1_data/ --hemi lh --projfrac 0.5 --interp trilinear --o perfusion_CBF_to_T1_surf_gray.mgh --reg register.dat --noreshape --cortex --surfreg sphere.reg
Step (3) takes the CBF map to cortical ribbon and hopefully now I can run mri_segstats and extract CBF values and thickness + do a group comparison using mri_glmfit. In order to do the latter, I have to map the perfusion maps to fsaverage. Can I then simply set --ref to fsaverage in step 3 or what is the correct call to take CBF maps to fsaverage in (3) ?
Second, Do i have to pass --surf argument pial or white in step 3? What is the correct way to do this?
Third, The statistical comparisons on thickness were done using 10 mm smoothing using -qcache argument in recon-all.I want to generate several smoothing of CBF maps (5-25 in steps of 5 mm) In the call to mri_vol2surf I can set --surf-fwhm 5 and my understanding is it will smooth the output surface 5 mm. Is that right? or should I call mri_fwhm on the output returned from step 3.
Thank you for your suggestions. I will really appreciate your thoughts.
Regards -VM