PET GTV pipeline

This is a pipeline made for $Citation needed

Installation

Ensure you have podman and Nvidia container toolkit installed. Contact repo owner for docker image

Start with: git clone https://github.com/Rigshospitalet-KFNM/HeadNeckPETGTV --recurse-submodules

Install virtual environment

python3 -m venv venv source venv/bin/activate pip install submodules/dicom_node

Then install dcm2niix and niftyreg found in the submodules

Environment variables

This pipeline is controlled by environment variables, set them using

export NAME=VALUE

• PIPELINE_ARCHIVE_PATH - Path, where the dicom node stores images temporary
• PIPELINE_WORKING_PATH - Path, where the dicom node executes the process function
• PIPELINE_LOG_PATH - Path, to the log file
• PIPELINE_DCM2NIIX - Path, to the dcm2niix executable
• PIPELINE_RESAMPLE - Path, to the reg_resample executable
• PIPELINE_AE_TITLE - Optional List of str, comma seperated, spaces are stripped. Accepted AE title, if present only ae titles in list is accepted
• PIPELINE_SEGMENTATION_ARCHIVE - Optional Path of directory, if present segmentation are saved to this directory.