发表机构
Chinese Academy of Medical Sciences; University of Cambridge（中国医学科学院; 剑桥大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出TMEvolve框架，将全切片图像建模为动态肿瘤微环境场，通过反应-扩散启发的演化过程整合区域扩散与边界通量，在六个数据集上优于现有方法。
AI 中文摘要
由于全切片图像（WSIs）具有千兆像素级别的特性，弱监督的WSI分析通常被表述为多实例学习（MIL）问题，其中补丁级特征被聚合为切片级表示。然而，诊断和预后证据往往来源于空间上连贯的肿瘤微环境区域及其相互作用，而非孤立的补丁。现有的基于补丁或静态区域的方法通常忽略了组织区域应如何自适应形成，并随后通过跨异质边界的微环境相互作用进行演化。在本文中，我们提出了概念引导的肿瘤微环境演化（TMEvolve），一个受反应-扩散启发的框架，将WSIs建模为离散补丁图上的潜在肿瘤微环境场。TMEvolve将这一观点实例化为一个可学习的、在补丁邻域上的图离散化演化过程。它首先形成自适应的软组织区域作为连贯的微环境单元，然后通过两种互补的局部动力学进行伪时间演化：区域内扩散，稳定连贯组织隔室内的潜在状态；以及概念引导的边界通量，在异质区域界面间传播视觉特征信号和语言派生的概念信号。演化的微环境区域最终被聚合用于切片级预测。我们在六个数据集上评估了TMEvolve，涵盖三个弱监督WSI任务：生存预测、基因表达预测和组织学亚型分类。TMEvolve在代表性的MIL方法、病理学基础模型和概念引导基线之上持续改进。消融研究和可视化进一步支持了TMEvolve的有效性和可解释性，强调了动态区域建模和边界交互的价值。
英文摘要
Due to the gigapixel-scale nature of whole-slide images (WSIs), weakly supervised WSI analysis is commonly formulated as a multiple instance learning (MIL) problem, where patch-level features are aggregated into slide-level representations. However, diagnostic and prognostic evidence often arises from spatially coherent tumor microenvironment regions and their interactions, rather than isolated patches alone. Existing patch-level or static region-based methods usually overlook how tissue regions should be adaptively formed and subsequently evolved through microenvironment interactions across heterogeneous boundaries. In this paper, we propose Concept-Guided Tumor Microenvironment Evolution (TMEvolve), a reaction-diffusion-inspired framework that models WSIs as latent tumor microenvironment fields over discrete patch graphs. TMEvolve instantiates this view as a learnable graph-discretized evolution process over patch neighborhoods. It first forms adaptive soft tissue regions as coherent microenvironment units, then performs pseudo-time evolution through two complementary local dynamics: intra-region diffusion, which stabilizes latent states within coherent tissue compartments, and concept-guided boundary flux, which propagates visual feature signals and language-derived concept signals across heterogeneous region interfaces. The evolved microenvironment regions are finally aggregated for slide-level prediction. We evaluate TMEvolve on six datasets across three weakly supervised WSI tasks: survival prediction, gene expression prediction, and histological subtype classification. TMEvolve consistently improves over representative MIL methods, pathology foundation models, and concept-guided baselines. Ablation studies and visualizations further support the effectiveness and interpretability of TMEvolve, highlighting the value of dynamic region modeling and boundary interaction.
CommentsAccepted at NeurIPS 2026
