On Sep 15, 2020, at 12:59 PM, Douglas N. Greve <dgreve@mgh.harvard.edu> wrote:

Both do both ways. I would recommend not using the --init-rr option for BOLD images

On 9/15/2020 11:42 AM, Wenzhen Zhao wrote:

        External Email - Use Caution        

Dear Douglas, 

Thank you for your quick response. I just have one more question. 

Does the “both-way” transformation only done through mri_coreg command that is embedded in preproc-sess command? Or does “bbregister —init-rr “ rigid transformation also does both-way transformation? 

Best, 

Wenzhen

On Sep 15, 2020, at 11:34 AM, Douglas N. Greve <dgreve@mgh.harvard.edu> wrote:

On 9/15/2020 11:26 AM, Wenzhen Zhao wrote:

        External Email - Use Caution        

Dear freesurfer developers, 

I’m attempting to analyze my resting -state fMRI data, and I followed procedures on your functional connectivity page. When I tried to run preprocessing with my data, but it turns out that default mri_coreg did not register my fMRI template volume onto anatomical scan well. So, I went on to register fMRI to anatomical space on my own using bbregister with (init-rr method - rigid transformation). I re-ran preprocessing without registration (-noreg) to correct motion and slice-timing. My commands looked like below: 

#bbregister —mov fMRI_template_volume —bold —s subject —init-rr —lta functional2anatomical.lta

#preproc-sess -s sess01 -fsd rest -stc siemens -noreg -surface fsaverage lhrh -surf-fwhm 5 -mni305-2mm -vol-fwhm 0 -per-run -force 

Then, I configured seeds (ex. Hippocampus), and ran following commands:

#fcseed-config -segid 17  -fcname left_hippocampus.dat -fsd rest -mean -cfg left_hippocampus.config

#fcseed-sess -s sess01 -cfg left_hippocampus.config 

It did run smoothly, but what I noticed from fcseed-sess script was it was transforming anatomical aseg mask into functional space. I am confused on this part, because during processing step, we only make registration file for transformation from functional space to anatomical space, but fcseed-sess command needs registration file that moves anatomical scan to functional scan.

Yes, it just inverts the transform to go in the opposite direction

# Convert segmentation to native functional space (##from fcseed-sess script##) 

      set cmd = (mri_label2vol --seg $seganat \

         --reg $reg --temp $template --fillthresh $FillThresh \

         --o $seg --pvf $pvf);

      echo $cmd | tee -a $LLF

Then, Later, during mkanalysis-sess step, the command seems to use functional 2 anatomical transformation. Could you please clarify which registration file it uses to process each step that I talked about above? Do I also need to make anatomical to functional transformation registration?

No, once it has a transformation in one direction, it has it in both directions

I deeply appreciate your help. 

Sincerely, 

Wenzhen 

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

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

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

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