---
datasets:
- ODELIA-AI/ODELIA-Challenge-2025
language:
- en
metrics:
- roc_auc
pipeline_tag: image-classification
tags:
- breast
- cancer
- mri
- odelia
- resnet
- 3d-classification
extra_gated_prompt: >-
  ### Model Usage Agreement

  By accessing or using this model (the "Model"), you acknowledge and agree
  to the following terms:

  1. The Model is provided for non-commercial academic and research use only.
  2. The Model must not be used for diagnosis, treatment, clinical decision
     making, or any application involving real-patient care.
  3. Do not identify, re-identify, or deanonymize people whose data may have
     contributed to the training or evaluation data.
  4. Publications or derivative research must acknowledge the ODELIA
     consortium and cite the Pimed authors and source repository.
  5. You are responsible for validating outputs in your research context.
     The Model is provided as-is, without warranties.

  This gated access does not grant rights beyond the license and permissions
  stated in this model card. Third-party components retain their own licenses.
---

# ODELIA Pimed Model

Three-class 3D breast MRI classification model contributed by the Pimed team
to the ODELIA Challenge 2025.

The source repository is
[dnschouten/odelia_submission_pimed](https://github.com/dnschouten/odelia_submission_pimed).

## Method

The corresponding implementation in the ODELIA project uses a 3D ResNet-18
backbone with one input channel and three output classes. It wraps the model
for subject-level classification of ODELIA MRI volumes.

The exact preprocessing, checkpoint configuration, and evaluation procedure
must be taken from the authors' source repository and the files uploaded with
this model. Do not infer clinical performance from the architecture alone.

## Intended use and limitations

This release is for research and benchmarking only. It is not a medical
device, is not clinically validated, and must not be used for diagnosis,
prognosis, treatment, triage, or other clinical decisions. Performance may
vary across institutions, scanners, acquisition protocols, and populations.

## Setup

Use the source repository for the authoritative setup and inference commands:

```bash
git clone https://github.com/dnschouten/odelia_submission_pimed.git
cd odelia_submission_pimed
```

## License and attribution

The implementation uses PyTorch/MONAI components. Their applicable licenses,
notices, and any pretrained-weight terms must be preserved and reviewed before
redistribution.

Please cite or acknowledge:

- ODELIA Challenge 2025 and the ODELIA consortium.
- The Pimed authors: Daan Schouten, Jeong Hoon Lee, Mirabela Rusu and [source repository](https://github.com/dnschouten/odelia_submission_pimed).
- The ODELIA paper: [arXiv:2506.00474](https://arxiv.org/abs/2506.00474).

For questions regarding model architecture, training, or permissions, please contact **Daan Schouten**.