Final Submission Requirements¶

Open source code | Deadline 01.10.2026¶

Winning teams are required to have their code open source before the award ceremony.

Algorithm description via form | Deadline 31.08.2026¶

Please answer the questions in this form.

Algorithm description in LNCS format | Deadline 31.08.2026¶

For the final test phase, submit a short paper (6-11 pages) reporting the details of your methods in LNCS format as a PDF. Attaching this file as a supplementary file along with your submission is compulsory.

Participant may submit one report covering two or more tasks if they wish. In that case please attach the report to each submission.

The template below outlines the key elements expected for a comprehensive description of an algorithm submitted to the DoseRAD2026 Grand Challenge.

Organizers reserve the right to exclude submissions lacking any of these reporting elements.

1. Title¶

• Specify the category of architecture used.

2. Abstract¶

Max 250 words for a single task, 400 words for a paper covering more than one task.

The abstract you provide will be used directly by the organizers as part of the challenge report. Submitting your method means you allow the organizers to use your descriptions for the future publication.

• Some example abstracts are provided in section 3 of arxiv.org/abs/2403.08447; please use these as inspiration and adapt them to your case.
• Briefly outline the methodology, including:
  • Method architecture and configuration
  • Whether the same method was used for all anatomical locations or tasks
  • Key techniques: highlight any specific techniques employed within the method
  • Loss function(s)
  • Optimizer and learning rate (including scheduling, if applicable)
  • Data preprocessing (in training, and in testing if performed) and augmentation
  • Image input size
  • Post-processing
  • Best model strategy: briefly explain how the final model was selected based on validation performance
• Mention any key results achieved (optional).

3. Introduction¶

• Provide the scientific and clinical background motivating the chosen methodology design.

4. Methods¶

Elaborate on all details mentioned in the abstract, following the structure from 4.1 to 4.4.

4.1 Data

• Specify the data subset used for hyperparameter optimization.
• Consider including an optional flow diagram illustrating data processing steps.
• If you used any additional public data (outside the DoseRAD2026 dataset) for training, please provide detailed information on this.

4.2 Model

• Provide a detailed description of the algorithm/model, including architecture, layers, and connections.
• Report the total number of parameters.
• List the software libraries, frameworks, and packages used.
• Explain the initialization of model parameters.
• Clearly indicate whether you used or fine-tuned a pre-trained model, and include a link to the corresponding repository.

4.3 Training

• Detail the training approach, including specific data augmentation techniques employed.
• Specify the hyperparameters used and their optimization methods.
• Describe the criteria for selecting the final model.
• If applicable, explain any ensembling techniques used.

4.4 Evaluation

• List the metrics used to assess model performance, including runtime. Please try to make use of the evaluation metric code provided with the challenge.
• Describe the statistical measures employed for significance and uncertainty (e.g. confidence intervals).
• Explain any methods used for model explainability or interpretability.

5. Results¶

• Report performance metrics for the optimized model(s) on specified dataset partitions (if used) and on the validation set.
• Analyze any output with poor performance.
• Include run-time performance on your hardware.

6. Discussion¶

• Discuss the limitations of the study, including potential bias and generalizability concerns.

7. Author contributions¶

• For transparency, we require corresponding authors to provide co-author contributions to the manuscript using the relevant CRediT roles. The CRediT taxonomy includes 14 different roles describing each contributor's specific contribution to the scholarly output.

8. Other information¶

• Acknowledge any sources of funding and collaborators.