Graph Mamba for Efficient Whole Slide Image Understanding

2025-05-23

AI Summary

To address scalability and computational efficiency bottlenecks arising from the ultra-high resolution, large size, and complex spatial relationships inherent in whole-slide images (WSIs), this paper proposes WSI-GMamba—a novel framework integrating graph neural networks (GNNs) with the Mamba state space model (SSM). We introduce the GMamba module, the first to synergistically combine graph message passing, graph-scanning flattening, and bidirectional SSM modeling—marking the inaugural application of Mamba to WSI-based multiple instance learning (MIL). Our method achieves Transformer-level classification accuracy while reducing inference FLOPs by 7×. Extensive experiments on multiple public WSI benchmark datasets demonstrate state-of-the-art (SOTA) performance, significantly enhancing both analytical efficiency and deployment feasibility for large-scale digital pathology applications.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Computer Vision: Multi-modal Vision
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

Whole Slide Images (WSIs) in histopathology present a significant challenge for large-scale medical image analysis due to their high resolution, large size, and complex tile relationships. Existing Multiple Instance Learning (MIL) methods, such as Graph Neural Networks (GNNs) and Transformer-based models, face limitations in scalability and computational cost. To bridge this gap, we propose the WSI-GMamba framework, which synergistically combines the relational modeling strengths of GNNs with the efficiency of Mamba, the State Space Model designed for sequence learning. The proposed GMamba block integrates Message Passing, Graph Scanning&Flattening, and feature aggregation via a Bidirectional State Space Model (Bi-SSM), achieving Transformer-level performance with 7* fewer FLOPs. By leveraging the complementary strengths of lightweight GNNs and Mamba, the WSI-GMamba framework delivers a scalable solution for large-scale WSI analysis, offering both high accuracy and computational efficiency for slide-level classification.

Problem

• Addresses high-resolution WSI analysis challenges
• Overcomes scalability limits of GNNs and Transformers
• Enhances efficiency in large-scale medical image processing

Innovation

• Combines GNNs with Mamba for efficiency
• Uses Bidirectional State Space Model (Bi-SSM)
• Achieves high accuracy with fewer FLOPs