GraphMMP: A Graph Neural Network Model with Mutual Information and Global Fusion for Multimodal Medical Prognosis

2025-08-24

AI Summary

To address the challenges of modeling heterogeneous modality interactions and insufficient joint capture of local and global dependencies in multimodal medical prognosis, this paper proposes a two-stage graph neural network framework. First, a cross-modal feature graph is constructed based on mutual information to explicitly model latent inter-modal relationships. Second, a Mamba-driven global fusion module is designed to efficiently integrate graph-structured representations, enabling long-range dependency modeling and multi-scale feature aggregation. Evaluated on the liver disease prognosis and METABRIC datasets, the method achieves significant performance gains over state-of-the-art approaches. Results demonstrate its effectiveness in deep association modeling of heterogeneous multimodal medical data and its strong generalization capability for end-to-end prognostic prediction.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

In the field of multimodal medical data analysis, leveraging diverse types of data and understanding their hidden relationships continues to be a research focus. The main challenges lie in effectively modeling the complex interactions between heterogeneous data modalities with distinct characteristics while capturing both local and global dependencies across modalities. To address these challenges, this paper presents a two-stage multimodal prognosis model, GraphMMP, which is based on graph neural networks. The proposed model constructs feature graphs using mutual information and features a global fusion module built on Mamba, which significantly boosts prognosis performance. Empirical results show that GraphMMP surpasses existing methods on datasets related to liver prognosis and the METABRIC study, demonstrating its effectiveness in multimodal medical prognosis tasks.

Problem

• Modeling complex interactions between heterogeneous medical data modalities
• Capturing local and global dependencies across multimodal medical data
• Improving prognosis performance through mutual information and global fusion

Innovation

• Graph neural networks with mutual information
• Global fusion module using Mamba architecture
• Two-stage multimodal prognosis modeling approach