DCMIL通过渐进式学习，从高分辨率病理图中自动预测癌症预后。
DCMIL: A Progressive Representation Learning of Whole Slide Images for Cancer Prognosis Analysis
- 采用双课程对比多实例学习，逐步挖掘不同放大倍数下的细粒度特征
- 在12种癌症、5954例患者数据上超越传统模型，准确率显著提升
- 无需密集标注，可识别关键病变区域，适合临床研究与生物发现
计算病理学在利用全幻灯片图像（WSIs）量化形态异质性、构建客观癌症预后模型方面展现出巨大潜力。然而，海量的十亿像素级输入带来计算瓶颈，且密集人工标注稀缺。现有方法常忽略多倍率WSI中的细粒度信息及肿瘤微环境差异。本文提出一种由易到难的渐进式表征学习方法——双课程对比多实例学习（DCMIL），可高效处理WSI进行癌症预后分析。该模型不依赖密集标注，能直接将十亿像素级WSI转化为预后预测。在12种癌症类型（5,954名患者，1254万张切片）上的大量实验表明，DCMIL优于标准的基于WSI的预后模型。此外，该方法能识别与预后相关的关键区域，提供稳健的实例不确定性估计，并捕捉正常与肿瘤组织间的形态差异，有望揭示新生物学机制。所有代码已公开于https://github.com/tuuuc/DCMIL。
原文摘要 · Abstract (English)
The burgeoning discipline of computational pathology shows promise in harnessing whole slide images (WSIs) to quantify morphological heterogeneity and develop objective prognostic modes for human cancers. However, progress is impeded by the computational bottleneck of gigapixel-size inputs and the scarcity of dense manual annotations. Current methods often overlook fine-grained information across multi-magnification WSIs and variations in tumor microenvironments. Here, we propose an easy-to-hard progressive representation learning, termed dual-curriculum contrastive multi-instance learning (DCMIL), to efficiently process WSIs for cancer prognosis. The model does not rely on dense annotations and enables the direct transformation of gigapixel-size WSIs into outcome predictions. Extensive experiments on twelve cancer types (5,954 patients, 12.54 million tiles) demonstrate that DCMIL outperforms standard WSI-based prognostic models. Additionally, DCMIL identifies fine-grained prognosis-salient regions, provides robust instance uncertainty estimation, and captures morphological differences between normal and tumor tissues, with the potential to generate new biological insights. All codes have been made publicly accessible at https://github.com/tuuuc/DCMIL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。