M-IDoL: Information Decomposition for Modality-Specific and Diverse Representation Learning in Medical Foundation Model
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909048165629952 |
|---|---|
| author | Liu, Yihang Yang, Longzhen Yang, Jiaxiong Wen, Ying He, Lianghua Shen, Heng Tao |
| author_facet | Liu, Yihang Yang, Longzhen Yang, Jiaxiong Wen, Ying He, Lianghua Shen, Heng Tao |
| contents | Medical foundation models (MFMs) aim to learn universal representations from multimodal medical images that can generalize effectively to diverse downstream clinical tasks. However, most existing MFMs suffer from information ambiguity that blends multimodal representations in a single embedding space, leading to the degradation of modality specificity and diversity. In this paper, we propose M-IDoL, a self-supervised MFM that introduces Information Decomposition for multimodal representation Learning via two objectives: i) maximizing inter-modality entropy by dispersing multimodal representations into separable Mixture-of-Experts (MoE) subspaces to achieve representation specificity across modalities; and ii) minimizing intra-modality uncertainty by performing fine-grained semantic discrimination within each MoE subspace to enrich representation diversity per modality. By pre-training on 1.15 million medical images, M-IDoL i) delivers superior generalization across 21 downstream clinical tasks, outperforming 20 foundation models on five imaging modalities (e.g., X-ray, fundus, OCT, dermoscopy and pathology), and ii) learns modality-specific and diverse representations, showing clearer separation of feature clusters across modalities and finer-grained feature discrimination within each modality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_08936 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | M-IDoL: Information Decomposition for Modality-Specific and Diverse Representation Learning in Medical Foundation Model Liu, Yihang Yang, Longzhen Yang, Jiaxiong Wen, Ying He, Lianghua Shen, Heng Tao Computer Vision and Pattern Recognition Medical foundation models (MFMs) aim to learn universal representations from multimodal medical images that can generalize effectively to diverse downstream clinical tasks. However, most existing MFMs suffer from information ambiguity that blends multimodal representations in a single embedding space, leading to the degradation of modality specificity and diversity. In this paper, we propose M-IDoL, a self-supervised MFM that introduces Information Decomposition for multimodal representation Learning via two objectives: i) maximizing inter-modality entropy by dispersing multimodal representations into separable Mixture-of-Experts (MoE) subspaces to achieve representation specificity across modalities; and ii) minimizing intra-modality uncertainty by performing fine-grained semantic discrimination within each MoE subspace to enrich representation diversity per modality. By pre-training on 1.15 million medical images, M-IDoL i) delivers superior generalization across 21 downstream clinical tasks, outperforming 20 foundation models on five imaging modalities (e.g., X-ray, fundus, OCT, dermoscopy and pathology), and ii) learns modality-specific and diverse representations, showing clearer separation of feature clusters across modalities and finer-grained feature discrimination within each modality. |
| title | M-IDoL: Information Decomposition for Modality-Specific and Diverse Representation Learning in Medical Foundation Model |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.08936 |