Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.17457 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911089685430272 |
|---|---|
| author | Lu, Jiaxuan Lin, Yuhui Shi, Junyan Yan, Fang Zhou, Dongzhan Gao, Yue Wang, Xiaosong |
| author_facet | Lu, Jiaxuan Lin, Yuhui Shi, Junyan Yan, Fang Zhou, Dongzhan Gao, Yue Wang, Xiaosong |
| contents | Whole Slide Images (WSIs) in histopathology pose a significant challenge for extensive medical image analysis due to their ultra-high resolution, massive scale, and intricate spatial relationships. Although existing Multiple Instance Learning (MIL) approaches like Graph Neural Networks (GNNs) and Transformers demonstrate strong instance-level modeling capabilities, they encounter constraints regarding scalability and computational expenses. To overcome these limitations, we introduce the WSI-HGMamba, a novel framework that unifies the high-order relational modeling capabilities of the Hypergraph Neural Networks (HGNNs) with the linear-time sequential modeling efficiency of the State Space Models. At the core of our design is the HGMamba block, which integrates message passing, hypergraph scanning & flattening, and bidirectional state space modeling (Bi-SSM), enabling the model to retain both relational and contextual cues while remaining computationally efficient. Compared to Transformer and Graph Transformer counterparts, WSI-HGMamba achieves superior performance with up to 7* reduction in FLOPs. Extensive experiments on multiple public and private WSI benchmarks demonstrate that our method provides a scalable, accurate, and efficient solution for slide-level understanding, making it a promising backbone for next-generation pathology AI systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17457 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Hypergraph Mamba for Efficient Whole Slide Image Understanding Lu, Jiaxuan Lin, Yuhui Shi, Junyan Yan, Fang Zhou, Dongzhan Gao, Yue Wang, Xiaosong Computer Vision and Pattern Recognition Artificial Intelligence Whole Slide Images (WSIs) in histopathology pose a significant challenge for extensive medical image analysis due to their ultra-high resolution, massive scale, and intricate spatial relationships. Although existing Multiple Instance Learning (MIL) approaches like Graph Neural Networks (GNNs) and Transformers demonstrate strong instance-level modeling capabilities, they encounter constraints regarding scalability and computational expenses. To overcome these limitations, we introduce the WSI-HGMamba, a novel framework that unifies the high-order relational modeling capabilities of the Hypergraph Neural Networks (HGNNs) with the linear-time sequential modeling efficiency of the State Space Models. At the core of our design is the HGMamba block, which integrates message passing, hypergraph scanning & flattening, and bidirectional state space modeling (Bi-SSM), enabling the model to retain both relational and contextual cues while remaining computationally efficient. Compared to Transformer and Graph Transformer counterparts, WSI-HGMamba achieves superior performance with up to 7* reduction in FLOPs. Extensive experiments on multiple public and private WSI benchmarks demonstrate that our method provides a scalable, accurate, and efficient solution for slide-level understanding, making it a promising backbone for next-generation pathology AI systems. |
| title | Hypergraph Mamba for Efficient Whole Slide Image Understanding |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2505.17457 |