arXiv cs.AI· Enhui Chai, Sicheng Chen, Tianyi Zhang, Chad Wong, Kecheng Huang, Zeyu Liu, Fei Xia·· 2 天前AI 评分20
BatMIL：面向全切片图像表示的双几何空间状态空间模型
BatMIL: Geometry-Embedding State Space Model for Whole-Slide Image Representation
AI 导读
BatMIL 提出一种混合双曲-欧氏几何表示，将全切片图像特征同时嵌入双几何空间，以兼顾组织层级结构与局部细胞形态。该方法采用线性复杂度的结构化状态空间序列模型建模数千 patch 的长程依赖，并用混合专家模块按特征动态路由 patch。在覆盖 6 种癌症的 7 个数据集上，BatMIL 性能优于 15 种 SOTA MIL 方法。
正文 · 原文
Abstract:Accurate analysis of histopathological images is critical for disease diagnosis and treatment planning. Whole-slide images (WSIs), which digitize tissue specimens at gigapixel resolution, are fundamental to this process but require aggregating thousands of patches for slide-level predictions. Multiple Instance Learning (MIL) tackles this challenge with a two-stage paradigm, decoupling tile-level embedding and slide-level prediction. However, most existing methods implicitly embed patch representations in homogeneous Euclidean spaces, overlooking the hierarchical organization and regional heterogeneity of pathological tissues. This limits current models' ability to capture global tissue architecture and fine-grained cellular morphology. To address this limitation, we introduce a hybrid hyperbolic-Euclidean representation that embeds WSI features in dual geometric spaces, enabling complementary modeling of hierarchical tissue structures and local morphological details. Building on this formulation, we develop BatMIL, an MIL method that leverages both geometric spaces. To model long-range dependencies among thousands of patches, we employ a structured state space sequence model backbone that encodes patch sequences with linear computational complexity. Furthermore, to account for regional heterogeneity, we introduce a mixture-of-experts module that dynamically routes each patch to specialized expert sub-networks based on its features, improving representational capacity while reducing redundant computation. Experiments on seven datasets spanning six cancer types demonstrate that BatMIL achieves outstanding performance against fifteen state-of-the-art MIL approaches, demonstrating that geometry-embedding representation learning offers a promising direction for computational pathology.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.05164 [cs.CV] |
| (or arXiv:2605.05164v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.05164 arXiv-issued DOI via DataCite |
Submission history
From: Sicheng Chen [view email]
[v1]
Wed, 6 May 2026 17:33:30 UTC (10,355 KB)
[v2]
Sat, 24 Oct 2026 19:36:39 UTC (22,742 KB)
来源：arXiv cs.AI · arxiv.org