search

PennLINC / xcp_d

Post-processing of fMRIPrep, NiBabies, and HCP outputs
https://xcp-d.readthedocs.io
BSD 3-Clause "New" or "Revised" License
78 stars 26 forks source link

Add data smoothness confound #843

Open tsalo opened 1 year ago

tsalo commented 1 year ago

Summary

@mattcieslak brought up the idea of estimating the smoothness of the data in scrum today, and using it as a confound across participants in analyses using ReHo or ALFF. I like this idea and think it's worth implementing in XCP-D.

One minor note: Matt mentioned that we need to subtract the mean BOLD image from the run's 4D data, or else the smoothness of the brain structures will affect 3dFWHMx's smoothness estimate.

tsalo commented 1 year ago

Should I instead use the residuals time series from denoising?