Filedump (dicom)

  • HM91
    • GCPDicom (garbage label)
  • HM92_1234
    • 1234 (proc ord id)
  • HM93
    • 1345
    • 3456
    • 9085
  • HM94 –> not included
    • GCPDicom
    • 7790483912084

First step: rename the sessions, keeping dicom format

sourcedata

  • HM91
    • procId345_ageDays00001
      • 72348921
        • 58493021
          • … .dcm
  • HM92
    • procId1234_ageDays01000
  • HM93
    • procId1345_ageDays01000
    • procId3456_ageDays01000
    • procId9085_ageDays02000

Second step: heudiconv, ideally after this step we can get rid of sourcedata (and dicom)

rawdata

  • sub-HM91
    • ses-procId345_ageDays00001
    • ses-GCPDicom_438129043_1473214123_1432145436_5425432523
      • anat (.nii .json)
      • dwi
  • HM92
    • procId1234_ageDays01000
  • HM93
    • procId1345_ageDays01000
    • procId3456_ageDays01000
    • procId9085_ageDays02000

Third step: CuBIDs - correctly identifying scans we know we want to analyze