组织学图像中癌症相关淋巴样聚集物的漫游指南：手动与基于深度学习的量化方法
Hitchhiker's guide to cancer-associated lymphoid aggregates in histology images: manual and deep learning-based quantification approaches
AI总结：
本研究针对癌症组织学图像中淋巴样聚集物量化的需求，提供了手动识别建议，开发了深度学习算法HookNet-TLS以克服手动分析的变异性，并给出了该算法的训练与部署指南，用于多种癌症中淋巴样聚集物的自动化客观量化。
摘要：
在癌症组织学图像中，对包含带有生发中心的三级淋巴结构在内的淋巴样聚集物进行量化，是开发预后及预测性组织生物标志物的一种极具前景的方法。本文针对常规病理工作流程（如苏木精-伊红染色）的组织切片，提供了淋巴样聚集物的识别建议。为克服手动图像分析固有的变异性（如主观决策、注意力持续时间限制），我们近期开发了一种名为HookNet-TLS的基于深度学习的算法，用于检测多种组织中的淋巴样聚集物与生发中心。本文还额外提供了使用手动标注图像训练并部署HookNet-TLS的指南，以实现对多种癌症类型中淋巴样聚集物的自动化、客观量化。
英文摘要：
Quantification of lymphoid aggregates including tertiary lymphoid structures with germinal centers in histology images of cancer is a promising approach for developing prognostic and predictive tissue biomarkers. In this article, we provide recommendations for identifying lymphoid aggregates in tissue sections from routine pathology workflows such as hematoxylin and eosin staining. To overcome the intrinsic variability associated with manual image analysis (such as subjective decision making, attention span), we recently developed a deep learning-based algorithm called HookNet-TLS to detect lymphoid aggregates and germinal centers in various tissues. Here, we additionally provide a guideline for using manually annotated images for training and implementing HookNet-TLS for automated and objective quantification of lymphoid aggregates in various cancer types.
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。
