DoseRAD2026 Challenge dataset: AI Accelerated Photon and Proton Dose Calculation for Radiotherapy
Summary: arXiv:2604.12778v1 Announce Type: cross
Abstract
Accurate dose calculation is essential in radiotherapy for precise tumor irradiation while sparing healthy tissue.
With the growing adoption of MRI-guided and real-time adaptive radiotherapy, fast and accurate dose calculation on CT and MRI is increasingly needed.
The DoseRAD2026 dataset and challenge provide a public benchmark of paired CT and MRI data with beam-level photon and proton Monte Carlo dose distributions for developing and evaluating advanced dose calculation methods.
Acquisition and Validation Methods
The DoseRAD2026 dataset comprises paired CT and MRI scans from 115 patients, including 75 for training and 40 for testing.
These patients were treated on an MRI-linac for thoracic or abdominal lesions, and the dataset was derived from the SynthRAD2025 dataset.
The pre-processing steps included:
- Deformable image registration
- Air-cavity correction
- Resampling
Ground-truth photon (6 MV) and proton dose distributions were computed using open-source Monte Carlo algorithms, resulting in an extensive dataset of
40,500 photon beams and 81,000 proton beamlets.
Data Format and Usage Notes
The dataset is organized into two main subsets: photon and proton, each containing paired CT-MRI images and beam-level dose distributions.
Additionally, JSON beam configuration files are provided to facilitate easy integration and analysis.
The files are offered in a compressed MetaImage (.mha) format, ensuring efficient storage and retrieval.
The dataset is released under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0), with training data set to become available in April 2026.
The test set will be withheld until March 2030 to maintain data integrity and allow for robust evaluation of methods developed using the training data.
Potential Applications
The DoseRAD2026 dataset supports a wide range of applications, including but not limited to:
- Benchmarking of fast dose calculation methods
- Beam-level dose estimation for both photon and proton therapy
- MRI-based dose calculation in MRI-guided workflows
- Real-time adaptive radiotherapy
By providing a public resource for researchers and practitioners, the DoseRAD2026 challenge aims to foster innovation and improve the accuracy of dose calculations in radiotherapy, ultimately enhancing patient outcomes and treatment efficacy.