crl/dicom-tools

Collection of tools for retrieving, formatting, organizing and converting DICOM images.

Step 1: Remove encoding python3 uncompress_dicoms.py dicomInputDirectory uncompressed/

Step 2: Sort the DICOM files python3 sort_dicoms.py uncompressed/ sorted/

Step 3: Convert each of the DICOM series to NIFTI python3 dicom_tree_to_nifti.py sorted/ converted/

These can be done from the command line with these steps:

• Remove JPEG2000 DICOM compression if needed:

docker run --rm -v "`pwd`":/data crl/dicom-tools uncompress_dicoms.py dicomdir dicomdir-uncompressed

• Sort the DICOM files into one series per directory:

docker run --rm -v "`pwd`":/data crl/dicom-tools sort_dicoms.py dicomdir-uncompressed dicomdir-sorted

• Convert the entire sorted DICOM directory tree to NIFTI format:

docker run --rm -v "`pwd`":/data crl/dicom-tools dicom_tree_to_nifti.py dicomdir-sorted dicomdir-converted

Getting the container at Boston Children's Hospital:

First authenticate to the internal BCH container registry and then run:

docker login ccts3.aws.chboston.org:5151 --username firstname.lastname --password glpat-access-token

To pull the container from the BCH gitlab container registry:

docker pull ccts3.aws.chboston.org:5151/computationalradiology/dicom-tools

Tag the local container with a shorter name:

docker tag ccts3.aws.chboston.org:5151/computationalradiology/dicom-tools crl/dicom-tools