Extracting WM and CSF

[JoffJones]

Hi there!

Thanks again for sharing this. I would really just like extract eroded WM and CSF signals from unsmoothed data that has had the AROMA noise ICs regressed out. However, I can't quite follow how you selected the WM and CSF ROIs. The current fmriprep segmentations are desc-aseg_dseg.nii.gz where 24 refers to CSF, 2 and 41 are cerebral WM, and 7 and 46 are cerebellar WM. I would appreciate your help!

Thanks,
Joff