arXiv 2505.046722026-02-05cs.CVq-bio.QM

Histo-Miner：基于深度学习的H&E全切片图像皮肤鳞状细胞癌组织特征提取流程

Histo-Miner: Deep learning based tissue features extraction pipeline from H&E whole slide images of cutaneous squamous cell carcinoma

Faculty of Mathematics and Natural Sciences, University of Cologne（数学与自然科学学院，科隆大学）；Excellence Cluster on Cellular Stress Responses in Aging-Associated Diseases (CECAD), University of Cologne（细胞应激反应与年龄相关疾病卓越集群（CECAD），科隆大学）；Department for Dermatology, University Hospital Cologne（皮肤科部门，科隆大学医院）；Department of Dermatology and Allergy, School of Medicine, Technical University of Munich（皮肤科与过敏科部门，医学院，慕尼黑技术大学）；Department of Dermatology, Dortmund Hospital gGmbH, University Witten/Herdecke（皮肤科部门，多特蒙德医院gGmbH，乌尔姆/赫尔德克大学）；Department of Dermatology, Venereology and Allergology, Helios St. Elisabeth Hospital Oberhausen, University Witten/Herdecke（皮肤科、性病科与过敏科部门，Helios圣埃利莎医院奥伯豪森，乌尔姆/赫尔德克大学）；Department of Dermatology and Allergology, University Hospital of the Paracelsus Medical University Salzburg（皮肤科与过敏科部门，帕拉塞尔医学大学萨尔茨堡大学医院）；Department of Dermatology and Allergology, University Hospital Bonn（皮肤科与过敏科部门，波恩大学医院）；Medical Clinic III for Oncology, Hematology, Immune-Oncology and Rheumatology, University Hospital Bonn (UKB)（肿瘤科、血液科、免疫肿瘤科和风湿科第三医疗部门，波恩大学医院（UKB））

Lucas Sancéré, Carina Lorenz, Doris Helbig, Oana-Diana Persa, Sonja Dengler, Alexander Kreuter, Martim Laimer, Roland Lang, Anne Fröhlich, Jennifer Landsberg, J… 展开作者

Lucas Sancéré, Carina Lorenz, Doris Helbig, Oana-Diana Persa, Sonja Dengler, Alexander Kreuter, Martim Laimer, Roland Lang, Anne Fröhlich, Jennifer Landsberg, Johannes Brägelmann, Katarzyna Bozek

备注：: 37 pages including supplement, 5 core figures. Version 2: change sections order, add new supplementary sections, minor text updates. Version 3: Author addition and update of author contributions, increase font on 2 figures, minor text updates
发表信息：: PLoS Comput. Biol., vol. 22, no. 1, p. e1013907, Jan. 2026
DOI：: 10.1371/journal.pcbi.1013907

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

Histo-Miner通过深度学习提取皮肤鳞状细胞癌全切片图像的组织特征，用于预测免疫治疗反应。

摘要：

近年来，数字病理学的进步使得能够对全切片图像（WSI）进行全面分析，利用高分辨率显微镜和计算能力。尽管取得了进展，但缺乏专门针对皮肤组织分析的标注数据集和开源流程。本文提出Histo-Miner，一个基于深度学习的皮肤WSI分析流程，并生成两个带有标注核和肿瘤区域的数据集。我们开发了该流程用于分析皮肤鳞状细胞癌（cSCC）患者样本。利用这两个数据集，包含47,392个标注细胞核和144个肿瘤分割的WSI，分别来自cSCC患者，Histo-Miner使用卷积神经网络和视觉Transformer进行核分割和分类以及肿瘤区域分割。训练模型的性能在多类Panoptic质量（mPQ）方面为0.569，核分类的宏平均F1为0.832，肿瘤区域分割的平均交并比（mIoU）为0-middle。从这些预测中，我们生成一个紧凑的特征向量，总结组织形态和细胞相互作用，可用于各种下游任务。在此，我们使用Histo-Miner根据45名患者术前WSI预测cSCC患者对免疫治疗的反应。Histo-Miner识别出淋巴细胞百分比、肿瘤附近粒细胞与淋巴细胞比值以及肿瘤中粒细胞与浆细胞之间的距离作为预测治疗反应的特征。这突显了Histo-Miner在临床相关场景中的适用性，提供了分类的直接解释和对潜在生物学的洞察。

英文摘要：

Recent advancements in digital pathology have enabled comprehensive analysis of Whole-Slide Images (WSI) from tissue samples, leveraging high-resolution microscopy and computational capabilities. Despite this progress, there is a lack of labeled datasets and open source pipelines specifically tailored for analysis of skin tissue. Here we propose Histo-Miner, a deep learning-based pipeline for analysis of skin WSIs and generate two datasets with labeled nuclei and tumor regions. We develop our pipeline for the analysis of patient samples of cutaneous squamous cell carcinoma (cSCC), a frequent non-melanoma skin cancer. Utilizing the two datasets, comprising 47,392 annotated cell nuclei and 144 tumor-segmented WSIs respectively, both from cSCC patients, Histo-Miner employs convolutional neural networks and vision transformers for nucleus segmentation and classification as well as tumor region segmentation. Performance of trained models positively compares to state of the art with multi-class Panoptic Quality (mPQ) of 0.569 for nucleus segmentation, macro-averaged F1 of 0.832 for nucleus classification and mean Intersection over Union (mIoU) of 0.907 for tumor region segmentation. From these predictions we generate a compact feature vector summarizing tissue morphology and cellular interactions, which can be used for various downstream tasks. Here, we use Histo-Miner to predict cSCC patient response to immunotherapy based on pre-treatment WSIs from 45 patients. Histo-Miner identifies percentages of lymphocytes, the granulocyte to lymphocyte ratio in tumor vicinity and the distances between granulocytes and plasma cells in tumors as predictive features for therapy response. This highlights the applicability of Histo-Miner to clinically relevant scenarios, providing direct interpretation of the classification and insights into the underlying biology.

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。