In the realm of Radiation Therapy Planning, the significance of segmentation cannot be overstated. Despite the existing availability of open-source models, there is a notable absence of comprehensive solutions that facilitate end-to-end segmentation for both organs at risk and target volumes. To address this gap, we present the A9T framework, leveraging the power of the nnUNet architecture. This module is specifically tailored for seamless integration into radiotherapy planning workflows.
- Seamlessly integrates with
DICOMimages, ensuring compatibility with standard medical imaging formats. - Deals with Structure Overlap by splitting models
- Predicts both
Organs-At-RiskandClinical Target Volumes - Works on multiple cancer sites:
TSPrime: Prostate Cancer PatientsTSGyne: Full Bladder Patients?
- Supports the execution of models in parallel using
multiprocessing - DRAW caters to the diverse needs of users by offering both automatic and manual segmentation options
- Database Integration for Monitoring and Analysing Workflows
For details about various commands, see documentation
Incoming
TODO