Steps in detail

What the script is doing

1. Sort dicom files according to series description

  • The script copies the raw dicom files to ${output_dir}/sourcedata/${subject}/ses-${session}.

  • Dicom files of for each series will be under a separate directory under

the target directory, {series_number}_{series_description}. - The script has been designed to also work with a dicom directory, where all dicom files are dumped under a single directory.

2. Dicom to BIDS using heudiconv

  • The U24 specific heuristic config file phantom_check/data/heuristic.py

is used. Output nifti files are saved under ${output_dir}/rawdata/${subject}/ses-${session}. - List of modality directories are

  • anat: contains T1w and T2w, both normalized and non-normalized.

  • dwi: contains diffusion weighted images (DWI). - PA acquisition (nifti, bval and bvec) - PA acquisition SBRef map - two AP b0 acquisitions (nifti, bval and bvec) - two AP SBRef maps

  • fmap: contains field maps. - AP and PA distortion maps before T1w scan - AP and PA distortion maps before first fMRI AP scan - AP and PA distortion maps before second fMRI AP scan

  • func: contains fMRI resting state scans. - First set of fMRI scans

    • AP and PA fMRI REST scans

    • SBRef volumes for each scans

    • Second set of fMRI scans

      • AP and PA fMRI REST scans

      • SBRef volumes for each scans

  • ignore: contains localizers and scout maps. - localizer - scout - alinged localizer

3. Quick quality check (QC) outputs

  • The script runs set of quality check pipelines to the data, and saves QC

outputs under ${output_dir}/derivatives/