Clinical brain MRI foundation models are going to cross the deployment gap soon 🧠
FOMO25 shows that SSL on heterogeneous scans can beat even in-domain supervised models in few-shot, cross-site settings.
This benchmark design feels like real progress.
Introducing FOMO300K — a large-scale, heterogeneous brain MRI dataset designed for self-supervised learning.
🧠 318,877 scans from 920 public datasets, spanning clinical & research MRI, multiple sequences, anatomies, and pathologies.
Deadline extended!
Interested in foundation models for brain MRI? You still have time to participate in the FOMO25 Challenge! Deadline has been extended until August 31, 2025.
To get started:
🌍 https://t.co/SvaqpaEYP7
💻 https://t.co/UvtaCwvIl7
🤗 https://t.co/qoXfbTxn7B
FOMO25 accepted at #MICCAI2025! 🚀
Excited to announce the first challenge at MICCAI focusing on the development of self-supervised pretraining of foundation models for brain MRI! 🧠
With access to a large-scale dataset, codebase, cash prices and multiple tracks.
Read more ⬇️