CoralBay: A Self-Supervised CT Foundation Model
For medical imaging researchers, CoralBay addresses the lack of effective 3D pre-training for CT scans, offering a general-purpose representation that outperforms 2D methods, though it is an incremental extension of DINO to 3D.
CoralBay introduces a self-supervised 3D CT foundation model using a hierarchical Swin backbone and self-distillation, achieving strong transfer performance across diverse radiological tasks. It also provides a public leaderboard for benchmarking volumetric representation learning.
Self-supervised learning has enabled large-scale pre-training on 2D natural images, producing general-purpose visual representations that transfer effectively across tasks. However, many medical imaging modalities, such as CT scans, are inherently three-dimensional and differ fundamentally from natural images in both structure and semantics. Volumetric modalities capture spatial continuity, organ anatomy, and intensity-based tissue properties (e.g., Hounsfield Units), which are not adequately modeled by 2D pre-training. To bridge this gap, we introduce CoralBay, a self-distillation framework that extends DINO by using a hierarchical 3D Swin backbone and applying self-distillation to concatenated multi-scale features, enabling data-efficient self-supervised learning of rich spatial representations that encode both global semantics and fine-grained local structure. As a result, CoralBay transfers effectively to a wide range of downstream radiological tasks, demonstrating strong and consistent performance across diverse anatomical targets. In addition, we contribute to the open-source \eva framework by introducing a public, reproducible 3D radiology leaderboard that unifies multiple datasets and establishes a standardized benchmark for evaluating volumetric representation learning methods.