用临床分级规则训练模型，从切片标签推断癌灶局部模式。
Weakly Supervised Instance-Level Gleason Pattern Estimation Using Primary and Secondary Labels
- 基于临床定义构建多实例学习框架，显式建模主次模式及其主导关系。
- 在SICAP-MIL数据集上优于现有方法，实现更准确的局部模式估计。
- 适合需要弱监督病理分析的研究者或医学影像算法开发者。
在前列腺癌组织病理学中，戈登评分由整张切片中出现最频繁（主要）和第二频繁（次要）的戈登模式决定。尽管临床实践中常规提供切片级别的主次标签，但实例级别的标注极少，导致图像块级别的学习困难。本文提出一种多实例学习（MIL）框架，从切片级别的主次标签中估计实例级别的戈登模式。该方法依据戈登评分的临床定义，通过聚合实例预测生成类别计数，并显式建模主要模式、次要模式及其主导关系。实验结果表明，该公式能够有效实现实例级别学习，在SICAP-MIL数据集上优于现有的MIL方法。
原文摘要 · Abstract (English)
In prostate cancer histopathology, the Gleason Score is determined by the most frequent (Primary) and second most frequent (Secondary) Gleason patterns within a whole-slide image. Although these slide-level labels are routinely available in clinical practice, instance-level Gleason annotations are rarely provided, making patch-level learning challenging. We propose a Multiple Instance Learning (MIL) framework that estimates instance-level Gleason patterns from slide-level Primary and Secondary labels. The proposed method formulates instance-level learning according to the clinical definition of the Gleason Score by aggregating instance predictions into class counts and explicitly modeling the Primary pattern, Secondary pattern, and their dominance. Experimental results demonstrate that the proposed formulation enables effective instance-level learning and outperforms existing MIL approaches on the SICAP-MIL dataset.
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。