Hi,
I have a question about preprocessing for surface-based analysis.
When I run afni_proc.py with the -surf block, I think it generates a script related to surface-based analysis.
It seems that the script generated by afni_proc.py mainly uses two functions involved in this process:@SUMA_AlignToExperiment and 3dVol2Surf .
If I use FreeSurfer and @SUMA_Make_Spec_FS , I can create several files such as spec files and cortical surfaces, such as pial and white matter surfaces, from the output of recon-all before running afni_proc.py.
Then, @SUMA_AlignToExperiment warps the FreeSurfer anatomical volume (e.g., ~SurfVol.nii ) to the experimental data – specifically the EPI/anatomical data used for registration in afni_proc.py.
I'm wondering about how this alignment process works, because I initially thought that not only ~SurfVol.nii, but also other files in the SUMA folder created by @SUMA_Make_Spec_FS, such as the spec file and surface files like pial and white matter surfaces, would need to be warped to the experimental data.
When I look at the @SUMA_AlignToExperiment website, it doesn’t appear to warp other files including spec files and cortical surfaces, such as pial and white matter surfaces.
So, I would like to know why I can still use the same spec file (e.g. std.141.~.*h.spec) found in the SUMA folder when I visualize surface-based GLM results.
The volume output by @SUMA_Make_Spec_FS contains the affine transformation to align those volumes (simple, since they are anats from the same subject) and that is applied automatically to the surfaces in question when they are read in to a SUMA program (such as 3dVol2Surf), and of course when that Alnd_Exp volume is used with -sv.
Using the run_suma script in the results directory, you will note that the surfaces are indeed aligned with this new Alnd_Exp volume, while those same surfaces stay aligned with SurfVol.nii in the SUMA directory.
Way back when that was decided, this approach seemed more efficient than creating an entire new set of surfaces.
Hi, rick
Thank you for your reply!
I would like to check whether I understand your reply correctly.
If I specify anatomical data that has been warped to the experimental data space by @SUMA_AlignToExperiment with the -sv option in the suma command, SUMA automatically aligns the surface files listed in the corresponding spec file (e.g., std.141.*h.spec), which are based on the ~SurfVol.nii, to the specified anatomical data.
This alignment between the warped anatomical data specified with -sv and surface files is simple because they are based on the same anatomical data - the difference is the coordinate space (Freesurfer or AFNI).
Yes, that is right, though rather than Freesurfer or AFNI space, I would consider them like anat1 space (used for FreeSurfer) and exp_anat space (the anat used in AlignToExp). Those original MRI anat coordinates should apply.
And for your comfort, please run afni and suma at the same time for those 2 cases:
and have them talk (afni and suma). The surfaces (from suma) should nicely align with the anats in the afni GUI. Have you done this before? If not, walk through the suma.pdf class handout that use the AFNI_data6 package. Sample commands for viewing start on slide 11.
Thank you for your reply and advice!
Yes, I have done that before, and the alignment between volumetric anatomical data and its surface seemed nice in the AFNI GUI and suma.
The
National Institute of Mental Health (NIMH) is part of the National Institutes of
Health (NIH), a component of the U.S. Department of Health and Human
Services.