arXiv 2504.173792025-04-25eess.IVcs.CV

一种用于数字病理学的空间感知多实例学习框架

A Spatially-Aware Multiple Instance Learning Framework for Digital Pathology

Department of Biomedical Engineering, Eindhoven University of Technology（生物医学工程系，埃因霍温理工大学）；Mathematics and Computer Science Department, Eindhoven University of Technology（数学与计算机科学系，埃因霍温理工大学）

Hassan Keshvarikhojasteh, Mihail Tifrea, Sibylle Hess, Josien P. W. Pluim, Mitko Veta

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对传统MIL忽略斑块空间交互的问题，提出GABMIL框架，在ABMIL中显式建模实例间依赖，保持计算效率，在肿瘤亚型分类上AUPRC提升7个百分点，Kappa提升5个百分点。

摘要：

多实例学习（MIL）是使用全切片图像（WSIs）进行病理学弱监督分类的一种有前景的方法。然而，传统的MIL方法，如基于注意力的深度多实例学习（ABMIL），通常忽略了对于病理诊断至关重要的斑块之间的空间交互。最近的进展，如基于Transformer的MIL（TransMIL），已经融入了空间上下文和斑块间关系。然而，目前尚不清楚在仅依赖多层感知器（MLPs）的ABMIL中，显式建模斑块关系是否能产生类似的性能提升。相比之下，TransMIL采用基于Transformer的层，引入了根本性的架构转变，但代价是计算复杂度显著增加。在这项工作中，我们通过整合交互感知表示来增强ABMIL框架，以解决这一问题。我们提出的模型，全局ABMIL（GABMIL），在保持计算效率的同时，显式捕获实例间的依赖关系。在两个公开可用的乳腺癌和肺癌肿瘤亚型分类数据集上的实验结果表明，与ABMIL相比，GABMIL在AUPRC上实现了高达7个百分点的提升，在Kappa分数上实现了5个百分点的提升，且几乎没有或完全没有额外的计算开销。这些发现强调了在MIL框架中纳入斑块交互的重要性。我们的代码可在\href{https://github.com/tueimage/GABMIL}{\texttt{GABMIL}}获取。

英文摘要：

Multiple instance learning (MIL) is a promising approach for weakly supervised classification in pathology using whole slide images (WSIs). However, conventional MIL methods such as Attention-Based Deep Multiple Instance Learning (ABMIL) typically disregard spatial interactions among patches that are crucial to pathological diagnosis. Recent advancements, such as Transformer based MIL (TransMIL), have incorporated spatial context and inter-patch relationships. However, it remains unclear whether explicitly modeling patch relationships yields similar performance gains in ABMIL, which relies solely on Multi-Layer Perceptrons (MLPs). In contrast, TransMIL employs Transformer-based layers, introducing a fundamental architectural shift at the cost of substantially increased computational complexity. In this work, we enhance the ABMIL framework by integrating interaction-aware representations to address this question. Our proposed model, Global ABMIL (GABMIL), explicitly captures inter-instance dependencies while preserving computational efficiency. Experimental results on two publicly available datasets for tumor subtyping in breast and lung cancers demonstrate that GABMIL achieves up to a 7 percentage point improvement in AUPRC and a 5 percentage point increase in the Kappa score over ABMIL, with minimal or no additional computational overhead. These findings underscore the importance of incorporating patch interactions within MIL frameworks. Our code is available at \href{https://github.com/tueimage/GABMIL}{\texttt{GABMIL}}.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。