Hi all, I used freesurfer to process my anatomical scans, followed by BBregister to align my resting state scan to the anatomical and finally afni to process the resting state fmri scans. After recon-all I used bbregister:
*bbregister -s ${sub}_FS --mov ${sub}.RS.nii --reg $sub_RS_register.dat --init-header --bold --frame $LAST --nearest --template-out $sub.RS.template.nii*
I used the following for manual adjustments (although none were needed!) *tkregister2 --mov $sub.RS.template.nii --reg $sub_RS_register.dat --surf*
The plan is to apply the reg file to the entire resting state timeseries ($sub.RS.nii).
My question is the following: 1) Does bbregister already move the last timepoint in $sub.RS.nii as well as $sub.RS.template.nii or does it just store the rigid transformation in the reg file (this is what I assume)? This is the reason I assume we supply reg into tkregister2 as I assume the transform is not applied to the nifti file header itself.
2) Assuming the transformations are not applied to the file itself, how would I go about doing this? I used the following based on previous posts:
*mri_vol2vol --mov $sub.RS.nii --reg $sub_RS_register.dat --fstarg --interp nearest --o $sub.RS.align.nii --precision short --no-resample * I used no-resample as I do not want to put the data into orig space. The plan is to align all of my timepoints to the last time point (same as $sub.RS.template.nii).
----Is --fstarg the best option (default orig.mgz vs --fstarg rawavg.mgz)? ----Is the transform stored in the header of $sub.RS.align.nii and not in the $sub.RS.template.nii file or $sub.RS.nii?
3) Using @SUMA_Make_Spec_FS did not correctly convert brain.mgz and aseg.mgz correctly when overlaid with the original $sub.SPGR.FS.mgz as well as $sub.RS.align.nii or $sub.RS.nii so I just used the following instead /mri directory:
*mri_convert -ot nii brain.mgz brain.nii*
*mri_convert -ot nii aseg.mgz aseg.nii*
*4) Following relates to viewing images created in AFNI correctly if anyone knows: * When I tried to use AFNI to view the original anatomical scan $sub.SPGR.FS.mgz and $sub.RS.align.nii it was not aligned (assume anatomical is similar to rawavg.mgz). When I load brain.nii and $sub.RS.align.nii into afni they are not overlapped perfectly as in tkregister2. Additionally I get an error about deobliquing the $sub.RS.align.nii file. Interestingly enough I do not get this error with the $sub.RS.template.nii so I am not sure if there is an error in applying the transform or if the template file has no transform info stored while the aligned file does. How would I apply the reg file to all timepoints without getting the oblique error or should I use the following command and is this correct:
*3dWarp -deoblique -gridset $sub.RS.nii -prefix $sub.RS.align.deobl.nii $sub.RS.align.nii*
Would i deoblique using the gridset of the original $sub.RS.nii, the aligned file $sub.RS.align.nii or just the template file (or does it not matter since the voxel dimensions are the same even though centers are different).
Sorry for the long email but thanks in advance for any advice / explanations.
Sincerely, Ajay