Physprep workflow

This standalone repository acts as use-case documentation for physiological data processing steps. The proposed workflow integrates community-based Ptyhon librairies such as phys2bids and neurokit2.

The repo is separated in three main modules, and provides a setp-by-step tutorial for each of them:

utils\

1. list_sub.py: list all the physiological files for a given subject (and a given session).
2. get_info.py: retrieve physiological files information.
3. match_acq_bids.py: match Acqknowledge files (.acq) with the fMRI Nifti files (.nii.gz).

preproc\

1. convert.py: use phys2bids to segment the acqknowledge files in runs following the BIDS format.
2. clean.py: implement functions to filter the physiological signals, and to remove the artifacts induced by the MRI.
3. process.py: build a processing pipeline based on clean.py functions.
4. quality.py: provide a summary of the quality of the processed signal.

visu\ 👷

Acqknowlegments

Thanks to the generous data donation from a subject of the Courtois-Neuromod project, research communities like PhysioPy will benefit from common data access to test and optimize their physio data preparation workflows, using BIDS format.