BrainDINO: A Brain MRI Foundation Model for Generalizable Clinical Representation Learning

ArXi:2604.27277v1 Announce Type: cross Brain MRI underpins a wide range of neuroscientific and clinical applications, yet most learning-based methods remain task-specific and require substantial labeled data. Here we show that a single self-supervised representation can generalize across heterogeneous brain MRI endpoints. We trained BrainDINO, a self-distilled foundation model, on approximately 6.6M unlabeled axial slices from 20 datasets encompassing broad variation in population, disease, and acquisition setting.