Deep learning-based survival prediction for multiple cancer types using histopathology images
Deep learning-based survival prediction for multiple cancer types using histopathology images
复制标题
基于深度学习的多种癌症类型的组织病理学图像生存预测
DOI：
10.1371/journal.pone.0233678
复制
发表时间：
2020-06-17
期刊：
影响因子：
3.7
通讯作者：
Stumpe, Martin C.
中科院分区：
文献类型：
--
作者：
Wulczyn, Ellery;Steiner, David F.;Stumpe, Martin C.
在癌症诊断时提供预后信息对于治疗和监测具有重要意义。尽管癌症分期、组织病理学评估、分子特征和临床变量可以提供有用的预后见解，但改善风险分层仍然是一个活跃的研究领域。我们开发了一个深度学习系统 (DLS)，用于预测癌症基因组图谱 (TCGA) 中 10 种癌症类型的疾病特异性生存率。我们使用了没有像素级注释的弱监督方法，并测试了三种不同的生存损失函数。 DLS 使用 3,664 个病例中的 9,086 张幻灯片进行开发，并使用 1,216 个病例中的 3,009 张幻灯片进行评估。在包括所有 10 种癌症的组合队列的多变量 Cox 回归分析中，DLS 与疾病特异性生存显着相关（风险比为 1.58，95% CI 1.28-1.70，p
Providing prognostic information at the time of cancer diagnosis has important implications for treatment and monitoring. Although cancer staging, histopathological assessment, molecular features, and clinical variables can provide useful prognostic insights, improving risk stratification remains an active research area. We developed a deep learning system (DLS) to predict disease specific survival across 10 cancer types from The Cancer Genome Atlas (TCGA). We used a weakly-supervised approach without pixel-level annotations, and tested three different survival loss functions. The DLS was developed using 9,086 slides from 3,664 cases and evaluated using 3,009 slides from 1,216 cases. In multivariable Cox regression analysis of the combined cohort including all 10 cancers, the DLS was significantly associated with disease specific survival (hazard ratio of 1.58, 95% CI 1.28-1.70, p