DICOMtoBIDS

How to use HeuDiConv (Heuristic DICOM Converter) to transform raw DICOM data into a BIDS dataset

Documentation and tutorials

https://github.com/nipy/heudiconv http://nipy.org/heudiconv/#10 https://www.youtube.com/watch?v=O1kZAuR7E00

*This how-to is adapted to use with docker (http://docker.com)

FIRST STEP: Create heuristic file (see convertKeys.py)

Copy the following code and paste it in a terminal:

docker run --rm -it
-v ~/../../../d/UGentData/fMRI/dicom:/data:ro
-v ~/../../../d/UGentData/fMRI/BIDS:/output
nipy/heudiconv:latest
-d /data/{subject}///*.IMA
-s '01'
-f convertall
-c none
-o /output

This will create a .heudiconv folder with a file called "convertall.py". You will have to adapt this file to the specifics of your study (see convert_retro.py file for an example).

At this point, move your new heuristic file to your script folder and delete the .heudiconv folder

Run the transformation with our heuristic file

for subject in {01..05}; do docker run --rm -it
-v ~/../../../d/UGentData/fMRI/dicom:/data:ro
-v ~/../../../d/UGentData/fMRI/BIDS:/output
nipy/heudiconv:latest
-d /data/{subject}///*.IMA
-s ${subject}
-f /data/scripts/convert_retro.py
-c dcm2niix -b
-o /output
--minmeta done