Giters

neurodatascience / nipoppy

Lightweight framework for neuroimaging-clinical data organization/processing

Home Page:https://nipoppy.readthedocs.io/en/latest/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Implement parallel processing for DICOM reorg/BIDS conversion/other workflows

michellewang opened this issue · comments

commented

The runscripts we have that use joblib are:

  • run_dicom_org.py
  • run_heudiconv.py
  • run_dcm2bids.py

Which translates into the following workflows:

  • DicomReorgWorkflow
  • BidsConversionRunner

Possibly other workflows that could benefit from parallel processing:

  • PipelineRunner (e.g. for users not running on an HPC)
  • PipelineTracker (can be slow)
  • DoughnutWorkflow maybe? I don't think that is very slow right now but it basically runs a loop to check files so shouldn't be too hard to implement