Early breast cancer detection is paramount and MRI increasingly supplements mammography for high-risk or dense-breasted women. Reading large 3D volumes demands significant radiologist time and is a barrier to the use of MRI in screening. To accelerate the development of AI tools for earlier, more accurate breast cancer detection in MRI, the ODELIA consortium is running the ODELIA Breast MRI Challenge 2025.
A dataset of 500 Breast MRI studies will be released for competing teams to use to train their models. These well-curated studies have been brought together from six centres across the ODELIA consortium and therefore presents a heterogeneous dataset of containing MRI scans with different parameters, protocols, and hardware. This heterogeneity is important to the training of well-generalised models which will have the widest real-world application.
The ODELIA Breast MRI Challenge 2025 has been accepted as a MICCAI challenge and the results will be presented at as part of the Deep-Breath workshop on 23 September 2025 at the conference organised at the Daejeon Convention Center, South Korea.
Eligible challenge winners will be invited to present their work either in person or online during the event.
It’s hoped that the challenge will also be a catalyst in the development of approaches and techniques which can be applied to the wider ODELIA project.
🔗 For more information on the challenge, visit https://odelia2025.grand-challenge.org/.
