M-IDoL: Information Decomposition for Modality-Specific and Diverse Representation Learning in Medical Foundation Model

• URL: http://arxiv.org/abs/2604.08936v1
• Date: Fri, 10 Apr 2026 04:06:11 GMT
• Title: M-IDoL: Information Decomposition for Modality-Specific and Diverse Representation Learning in Medical Foundation Model
• Abstract summary: We propose M-IDoL, a self-supervised underlinetextitMFM that introduces Information Decomposition for multimodal representation Learning.<n>By pre-training on 1.15 million medical images, M-IDoL i) delivers superior generalization across 21 downstream clinical tasks.
• Score: 53.96788246923603
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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 blend 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 \underline{\textit{M}}FM that introduces Information Decomposition for multimodal representation Learning via two objectives: i) maximize inter-modality entropy by dispersing multimodal representation into separable Mixture-of-Experts (MoE) subspaces to achieve representation specificity across modalities; and ii) minimize 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 cluster across modalities and finer-grained feature discrimination within each modality.

Related papers

• Discrete Diffusion Models with MLLMs for Unified Medical Multimodal Generation [30.047148886472637]
  We propose MeDiM, a medical discrete diffusion model that learns shared distributions across modalities without modality-specific components.<n>MeDiM unifies multiple generative tasks: translating between images and text, and jointly producing image-report pairs across domains in response to prompts.
  arXiv  Detail & Related papers  (2025-10-07T17:06:57Z)
• Multimodal Causal-Driven Representation Learning for Generalizable Medical Image Segmentation [56.52520416420957]
  We propose Multimodal Causal-Driven Representation Learning (MCDRL) to tackle domain generalization in medical image segmentation.<n>MCDRL consistently outperforms competing methods, yielding superior segmentation accuracy and exhibiting robust generalizability.
  arXiv  Detail & Related papers  (2025-08-07T03:41:41Z)
• Robust Multimodal Learning for Ophthalmic Disease Grading via Disentangled Representation [30.697291934309206]
  multimodal data is rare in real-world applications due to a lack of medical equipment and concerns about data privacy.<n>Traditional deep learning methods typically address these issues by learning representations in latent space.<n>Authors propose the Essence-Point and Disentangle Representation Learning (EDRL) strategy, which integrates a self-distillation mechanism into an end-to-end framework.
  arXiv  Detail & Related papers  (2025-03-07T10:58:38Z)
• Incomplete Modality Disentangled Representation for Ophthalmic Disease Grading and Diagnosis [16.95583564875497]
  We propose an Incomplete Modality Disentangled Representation (IMDR) strategy to disentangle features into explicit independent modal-common and modal-specific features.<n> Experiments on four multimodal datasets demonstrate that the proposed IMDR outperforms the state-of-the-art methods significantly.
  arXiv  Detail & Related papers  (2025-02-17T12:10:35Z)
• MIND: Modality-Informed Knowledge Distillation Framework for Multimodal Clinical Prediction Tasks [50.98856172702256]
  We propose the Modality-INformed knowledge Distillation (MIND) framework, a multimodal model compression approach.<n>MIND transfers knowledge from ensembles of pre-trained deep neural networks of varying sizes into a smaller multimodal student.<n>We evaluate MIND on binary and multilabel clinical prediction tasks using time series data and chest X-ray images.
  arXiv  Detail & Related papers  (2025-02-03T08:50:00Z)
• Multi-modal MRI Translation via Evidential Regression and Distribution Calibration [29.56726531611307]
  We propose a novel framework that reformulates multi-modal MRI translation as a multi-modal evidential regression problem with distribution calibration.<n>Our approach incorporates two key components: 1) an evidential regression module that estimates uncertainties from different source modalities and an explicit distribution mixture strategy for transparent multi-modal fusion, and 2) a distribution calibration mechanism that adapts to source-target mapping shifts.
  arXiv  Detail & Related papers  (2024-07-10T05:17:01Z)
• Enhancing Multimodal Unified Representations for Cross Modal Generalization [52.16653133604068]
  We propose Training-free Optimization of Codebook (TOC) and Fine and Coarse cross-modal Information Disentangling (FCID)<n>These methods refine the unified discrete representations from pretraining and perform fine- and coarse-grained information disentanglement tailored to the specific characteristics of each modality.
  arXiv  Detail & Related papers  (2024-03-08T09:16:47Z)
• Multi-task Paired Masking with Alignment Modeling for Medical Vision-Language Pre-training [55.56609500764344]
  We propose a unified framework based on Multi-task Paired Masking with Alignment (MPMA) to integrate the cross-modal alignment task into the joint image-text reconstruction framework. We also introduce a Memory-Augmented Cross-Modal Fusion (MA-CMF) module to fully integrate visual information to assist report reconstruction.
  arXiv  Detail & Related papers  (2023-05-13T13:53:48Z)
• Robust Multimodal Brain Tumor Segmentation via Feature Disentanglement and Gated Fusion [71.87627318863612]
  We propose a novel multimodal segmentation framework which is robust to the absence of imaging modalities. Our network uses feature disentanglement to decompose the input modalities into the modality-specific appearance code. We validate our method on the important yet challenging multimodal brain tumor segmentation task with the BRATS challenge dataset.
  arXiv  Detail & Related papers  (2020-02-22T14:32:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.