We are preprocessing our data externally using FSL and importing the preprocessed data into CONN. As one part of this, we have existing WM/CSF subject-specific structural masks  (not TPMs) such as 'sub-XX_whitematter.nii', and 'sub-XX_cerebrospinalfluid.nii'. We would like to use these masks to extract the CSF/WM timecourses for denoising. Is there a way to do that using batchscript? Do we need to add any prefixes before importing them as ROI in the Setup.ROI? Our imported structural and functional data have undergone all the preprocessing steps in FSL, except for slice-time correction. Our primary goal is to perfrom denoising and functional connectivity analysis in CONN on this preprocessed data. 

 

If we don't run any preproc segmentation after importing the data, the WM/CSF variables in Denoising look wrong/nonexistnt. We tried running functional segmentation, but got the error message "mean functional file not found" (details below). How would we create the mean functional file? There doesn't seem to be a preprocessing step for that. Or is this all the wrong approach, and we need to be doing something different to get out those aCompCor variables from WM/CSF tissue masks?