MAPI-GNN: Multi-Activation Plane Interaction Graph Neural Network for Multimodal Medical Diagnosis

2025-12-22

AI Summary

Existing graph neural networks for multimodal medical diagnosis rely on a single static graph, failing to capture patient-specific pathological relationships and thereby limiting diagnostic accuracy and personalization. To address this, we propose a dynamic multi-activated planar graph modeling framework tailored for multimodal medical diagnosis. First, patient-specific graph structures are generated via semantic-decoupled feature subspaces. Second, a graph-aware multi-dimensional discriminator and a relational fusion engine are designed to enable fine-grained pathological relationship modeling and context-adaptive aggregation. Our work introduces, for the first time, the “multi-activated planar graph” generation paradigm, overcoming inherent limitations of static graph modeling. Evaluated on two cross-modal clinical tasks involving over 1,300 patients, our method achieves an average diagnostic accuracy improvement of 4.2%, significantly outperforming state-of-the-art approaches.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Graph neural networks are increasingly applied to multimodal medical diagnosis for their inherent relational modeling capabilities. However, their efficacy is often compromised by the prevailing reliance on a single, static graph built from indiscriminate features, hindering the ability to model patient-specific pathological relationships. To this end, the proposed Multi-Activation Plane Interaction Graph Neural Network (MAPI-GNN) reconstructs this single-graph paradigm by learning a multifaceted graph profile from semantically disentangled feature subspaces. The framework first uncovers latent graph-aware patterns via a multi-dimensional discriminator; these patterns then guide the dynamic construction of a stack of activation graphs; and this multifaceted profile is finally aggregated and contextualized by a relational fusion engine for a robust diagnosis. Extensive experiments on two diverse tasks, comprising over 1300 patient samples, demonstrate that MAPI-GNN significantly outperforms state-of-the-art methods.

Problem

• Enhances multimodal medical diagnosis with dynamic patient-specific graphs
• Overcomes single static graph limitations in modeling pathological relationships
• Learns multifaceted graph profiles from semantically disentangled feature subspaces

Innovation

• Learns multifaceted graph profile from disentangled feature subspaces
• Dynamically constructs activation graphs guided by latent patterns
• Aggregates multifaceted profile via relational fusion engine