search

PennLINC / xcpEngine

Official public repository for the XCP Engine. This tool is deprecated in favor of XCP-D and ASLPrep.
MIT License
66 stars 42 forks source link

confound2/mc relRMS.1D == fd.1D #397

Closed dominikkraft closed 3 years ago

dominikkraft commented 4 years ago

Within the confound module, it appears that utils/generate_confmat.R reads the framewise displacement from the fmriprep confound table and saves it as subject_relRMS.1D. But I think this should refer to Jenkinsons RMS displacement.

Code chunk:

{XCPEDIR}/utils/generate_confmat.R -i /data/hippocampus/FLEX/FLEX_rest_FINAL/BIDS/test_denoised/sub-vfo297/prestats/sub-vfo297_fmriconf.tsv -j fd -o /data/hippocampus/FLEX/FLEX_rest_FINAL/BIDS/test_denoised/sub-vfo297/confound2/mc/sub-vfo297_relRMS.1D

The FD _fd.1D calculated within /utils/fd.R appears to be identical to RMS.1D , too.

Given that fmriprep now outputs rmsd one could integrate it here (under RMS.1D) and avoid confusion and allow for better stream lining with fmriprep versions 20.X.X and newer.

Best, Dominik

a3sha2 commented 4 years ago

thank you @dominikkraft