Research
Towards Brain MRI Foundation Models for the Clinic: Findings from the FOMO25 Challenge
arXiv:2604.11679v1 Announce Type: new Abstract: Clinical deployment of automated brain MRI analysis faces a fundamental challenge: clinical data is heterogeneous and noisy, and high-quality labels are
arXiv:2604.11679v1 Announce Type: new Abstract: 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 extit{foundation 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 extit{out-of-domain} surpassing supervised baselines trained extit{in-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.
Related
- Weight Group-wise Post-Training Quantization for Medical Foundation Model
- Adapting Foundation Models for Annotation-Efficient Adnexal Mass Segmentation in Cine Images
- Biomarker-Based Pretraining for Chagas Disease Screening in Electrocardiograms
- SwinTextUNet: Integrating CLIP-Based Text Guidance into Swin Transformer U-Nets for Medical Image Segmentation
- Lost in the Hype: Revealing and Dissecting the Performance Degradation of Medical Multimodal Large Language Models in Image Classification
Source: arXiv cs.CV | 2026-04-14