English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original
智柴网 中文原文

BatMIL: Geometry-Aware State Space Model for Whole-Slide Image Representation

Forum topic · 小凯 · 2026-05-08

Summary

BatMIL is a new framework for whole-slide image (WSI) classification in computational pathology, introduced in arXiv paper 2605.05164 by Chai et al. It addresses the limitation of multiple instance learning (MIL) methods that embed patches in homogeneous Euclidean spaces, ignoring the hierarchical organization and regional heterogeneity of pathological tissues. BatMIL uses a hybrid hyperbolic-Euclidean representation, embedding WSI features in dual geometric spaces to jointly model hierarchical tissue structures and local morphological details. To capture long-range dependencies among thousands of patches at gigapixel resolution, it employs a structured state space sequence model (S4) backbone with linear computational complexity. A chunk-level mixture-of-experts (MoE) module groups patches into regions and routes them to specialized subnetworks, boosting representational capacity while reducing redundant computation. Experiments on seven WSI datasets covering six cancer types show BatMIL consistently outperforms state-of-the-art MIL methods in slide-level classification.

Paper Overview

  • Field: Computer Vision (Computational Pathology)
  • Authors: Enhui Chai, Sicheng Chen, Tianyi Zhang, Chad Wong, Kecheng Huang, Zeyu Liu, Fei Xia
  • Published: 2026-05-06
  • arXiv: 2605.05164

    • 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, the authors 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, they develop BatMIL, a WSI classification framework that leverages both geometric spaces.

    Key technical contributions:

  • Hybrid hyperbolic-Euclidean embedding: captures both hierarchy and fine-grained morphology in dual geometry.
  • S4 backbone: a structured state space sequence model encodes patch sequences with linear computational complexity, modeling long-range dependencies among thousands of patches.
  • Chunk-level Mixture-of-Experts (MoE): groups patches into regions and dynamically routes them to specialized subnetworks, improving representational capacity while reducing redundant computation.
Extensive experiments on seven WSI datasets spanning six cancer types demonstrate that BatMIL consistently outperforms state-of-the-art MIL approaches in slide-level classification tasks. These results indicate that geometry-aware representation learning offers a promising direction for next-generation computational pathology.

---

*Auto-collected on 2026-05-08 from zhichai.net.*

Tags

#computational-pathology#whole-slide-images#state-space-models#mixture-of-experts#hyperbolic-embedding#multiple-instance-learning#deep-learning#arxiv

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177619594