fMRIPrep for longitudinal brain injury patients

1

Hi all,

I’m preprocessing a longitudinal brain injury dataset with two sessions per subject (ses-1 and ses-2, ~9 months). Because tissue recovery can be substantial across sessions, I’m unsure about the best way to organize FreeSurfer outputs and how to use a lesion mask in the preprocessing pipeline.

I noticed that the longitudinal workflow in fMRIPrep seems to create an average anatomical template across timepoints. Given that anatomy may truly change over 9 months, I’m wondering whether it is better way to run fMRIPrep separately for each session or treating each session independently, rather than using a longitudinal template approach.

Any recommendations or example workflows would be greatly appreciated.

Thanks!

2

Hi @lei_wei,

You can organize freesurfer in session specific folders, and pass those in for the FS subjects dir. Or you can run longitudinal freesurfer outside of fMRIPrep and pass in those inputs, additionally specifying a —fs-no-resume flag. If you can combine longitudinally, that makes analyses between time points safer. We cannot say which approach makes more sense without knowing more about the data.

Lesion mask instructions are here: Processing pipeline details — fmriprep version documentation

Best,
Steven