If you are combining the T1 images together in recon-all, then it is not possible to separate them in FSFAST. If you do not expect structural changes, then you can just register both fMRI sessions to the single anatomical. If you are expecting changes, then analyze the T1 images separately (in the longitudinal stream). This creates two subject names; use the appropriate one in each fmri session.

On 02/16/2018 02:30 AM, Ben Smith wrote:

(Thanks for your feedback to my previous inquiry - I've resolved this problem)

I'm running FsFastPreProc on a dataset with a number of subjects, where each subject has two sessions.

There's a separate structural image taken in each session. I want to make sure each subject's functional data is registered to the correct structural image. When running fsfast, setting up the file structure I can see clearly how FsFast uses the subjectname file to identify the right subject. It also seems easy enough to register multiple T1 images - just include multiple inputs for each subject to the *recon-all* command.

But how do I map the session to the appropriate T1 image in the structural data?

Also, if I were registering feat directories to structural using reg-feat2anat - how would I specify the session in that case?

Does this even matter? Perhaps FSFastPreProc is built to process images without having a T1 from the same session as each image.

If there's not an easy way to do this, would a suitable workaround be to create a separate anatomical subject folder for every subject*session?

Thanks very much for your help!

Best regards

Ben

---

Ben Smith, MSc. Ph.D. Candidate University of Southern California https://bjsmith.github.io/ w | +1 213-740 7851 m | +1 323-385 9349 skype | ben.smith.nz http://ben.smith.nz/

---

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