# Towards Brain MRI Foundation Models for the Clinic: Findings from the FOMO25 Challenge

arXiv: https://arxiv.org/abs/2604.11679 · HTML: https://mlsanity.com/paper/2604.11679/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 52 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Builds on (most cited first, 30 of 34)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Improved Baselines with Momentum Contrastive Learning](https://mlsanity.com/paper/2003.04297/) (2020)
- [SimMIM: A Simple Framework for Masked Image Modeling](https://mlsanity.com/paper/2111.09886/) (2021)
- [iBOT: Image BERT Pre-Training with Online Tokenizer](https://mlsanity.com/paper/2111.07832/) (2021)
- [Swin UNETR: Swin Transformers for Semantic Segmentation of Brain Tumors in MRI Images](https://mlsanity.com/paper/2201.01266/) (2022)
- [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](https://mlsanity.com/paper/2301.00808/) (2023)
- [TotalSegmentator: robust segmentation of 104 anatomical structures in CT images](https://mlsanity.com/paper/2208.05868/) (2022)
- [Self-Supervised Pre-Training of Swin Transformers for 3D Medical Image Analysis](https://mlsanity.com/paper/2111.14791/) (2021)
- [xLSTM: Extended Long Short-Term Memory](https://mlsanity.com/paper/2405.04517/) (2024)
- [nnU-Net Revisited: A Call for Rigorous Validation in 3D Medical Image Segmentation](https://mlsanity.com/paper/2404.09556/) (2024)
- [SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining](https://mlsanity.com/paper/2107.09559/) (2021)
- [UniFormer: Unifying Convolution and Self-attention for Visual Recognition](https://mlsanity.com/paper/2201.09450/) (2022)
- [Metrics reloaded: Recommendations for image analysis validation](https://mlsanity.com/paper/2206.01653/) (2022)
- [STU-Net: Scalable and Transferable Medical Image Segmentation Models Empowered by Large-Scale Supervised Pre-training](https://mlsanity.com/paper/2304.06716/) (2023)
- [Why rankings of biomedical image analysis competitions should be interpreted with care](https://mlsanity.com/paper/1806.02051/) (2018)
- [VoCo: A Simple-yet-Effective Volume Contrastive Learning Framework for 3D Medical Image Analysis](https://mlsanity.com/paper/2402.17300/) (2024)
- [Masked Image Modeling Advances 3D Medical Image Analysis](https://mlsanity.com/paper/2204.11716/) (2022)
- [BIAS: Transparent reporting of biomedical image analysis challenges](https://mlsanity.com/paper/1910.04071/) (2019)
- [The reliability of a deep learning model in clinical out-of-distribution MRI data: a multicohort study](https://mlsanity.com/paper/1911.00515/) (2019)
- [Revisiting MAE pre-training for 3D medical image segmentation](https://mlsanity.com/paper/2410.23132/) (2024)
- [An OpenMind for 3D medical vision self-supervised learning](https://mlsanity.com/paper/2412.17041/) (2024)
- [How Robust is Unsupervised Representation Learning to Distribution Shift?](https://mlsanity.com/paper/2206.08871/) (2022)
- [AMAES: Augmented Masked Autoencoder Pretraining on Public Brain MRI Data for 3D-Native Segmentation](https://mlsanity.com/paper/2408.00640/) (2024)
- [Curia: A Multi-Modal Foundation Model for Radiology](https://mlsanity.com/paper/2509.06830/) (2025)
- [A large-scale heterogeneous 3D magnetic resonance brain imaging dataset for self-supervised learning](https://mlsanity.com/paper/2506.14432/) (2025)

## Built on by (most cited first, 3 of 3)

- [Contrastive Joint-Embedding Prediction for Representation Learning in Structural MRI](https://mlsanity.com/paper/2607.11962/) (2026)
- [BrainNext: A General-Purpose Self-Supervised Foundation Model for Brain MRI Analysis](https://mlsanity.com/paper/2607.17782/) (2026)
- [NeuroCBIR: A Fast and Accurate Image Retrieval System for Whole-Brain and Region-Specific MRI](https://mlsanity.com/paper/2610.06502/) (2026)

## Abstract (arXiv)

Clinical deployment of automated brain MRI analysis faces a fundamental challenge: clinical data is heterogeneous and noisy, and high-quality labels are prohibitively costly to obtain. Self-supervised learning (SSL) can address this by leveraging the vast amounts of unlabeled data produced in clinical workflows to train robust textitfoundation models that adapt out-of-domain with minimal supervision. However, the development of foundation models for brain MRI has been limited by small pretraining datasets and in-domain benchmarking focused on high-quality, research-grade data. To address this gap, we organized the FOMO25 challenge as a satellite event at MICCAI 2025. FOMO25 provided participants with a large pretraining dataset, FOMO60K, and evaluated models on data sourced directly from clinical workflows in few-shot and out-of-domain settings. Tasks covered infarct classification, meningioma segmentation, and brain age regression, and considered both models trained on FOMO60K (method track) and any data (open track). Nineteen foundation models from sixteen teams were evaluated using a standardized containerized pipeline. Results show that (a) self-supervised pretraining improves generalization on clinical data under domain shift, with the strongest models trained textitout-of-domain surpassing supervised baselines trained textitin-domain. (b) No single pretraining objective benefits all tasks: MAE favors segmentation, hybrid reconstruction-contrastive objectives favor classification, and (c) strong performance was achieved by small pretrained models, and improvements from scaling model size and training duration did not yield reliable benefits.