个人中心| | 帮助
高级检索
当前位置： 首页 > 详情页

Domain generalization across tumor types, laboratories, and species - Insights from the 2022 edition of the Mitosis Domain Generalization Challenge

| 认领 | 导出 | |

文献详情

资源类型：
Pubmed体系：
作者：
Marc Aubreville[1] ; Nikolas Stathonikos[2] ; Taryn A. Donovan[3] ; Robert Klopfleisch[4] ; Jonas Ammeling [1] ; Jonathan Ganz [1] ; Frauke Wilm [56] ; Mitko Veta [7] ; Samir Jabari [8] ; Markus Eckstein [9] ; Jonas Annuscheit [10] ; Christian Krumnow [10] ; Engin Bozaba [11] ; Sercan Çayır [11] ; Hongyan Gu [12] ; Xiang ‘Anthony’ Chen [12] ; Mostafa Jahanifar [13] ; Adam Shephard [13] ; Satoshi Kondo [14] ; Satoshi Kasai [15] ; Sujatha Kotte [16] ; V.G. Saipradeep [16] ; Maxime W. Lafarge [17] ; Viktor H. Koelzer [17] ; Ziyue Wang [18] ; Yongbing Zhang [18] ; Sen Yang [19] ; Xiyue Wang [20] ; Katharina Breininger[6] ; Christof A. Bertram [21] ;
机构： [1]Technische Hochschule Ingolstadt, Ingolstadt, Germany [2]Pathology Department, UMC Utrecht, The Netherlands [3]Department of Anatomic Pathology, The Schwarzman Animal Medical Center, NY, USA [4]Institute of Veterinary Pathology, Freie Universität Berlin, Berlin, Germany [5]Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany [6]Department Artificial Intelligence in Biomedical Engineering, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany [7]Computational Pathology Group, Radboud UMC Nijmegen, The Netherlands [8]Institute of Neuropathology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany [9]Institute of Pathology, University Hospital Erlangen, Friedrich-Alexander-Universität Erlangen-Nünberg, Erlangen, Germany [10]University of Applied Sciences (HTW) Berlin, Berlin, Germany [11]Artificial Intelligence Research Team, Virasoft Corporation, NY, USA [12]University of California, Los Angeles, USA [13]University of Warwick, United Kingdom [14]Muroran Institute of Technology, Muroran, Japan [15]Niigata University of Health and Welfare, Niigata, Japan [16]TCS Research, Tata Consultancy Services Ltd, Hyderabad, India [17]Department of Pathology and Molecular Pathology, University Hospital Zurich, University of Zurich, Zurich, Switzerland [18]Harbin Institute of Technology, Shenzhen, China [19]College of Biomedical Engineering, Sichuan University, Chengdu, China [20]Department of Radiation Oncology, Stanford University School of Medicine, Palo Alto, USA [21]Institute of Pathology, University of Veterinary Medicine, Vienna, Austria
出处：
DOI： 10.1016/j.media.2024.103155
ISSN：

关键词： Domain generalization Histopathology Challenge Deep Learning Mitosis

摘要：
Recognition of mitotic figures in histologic tumor specimens is highly relevant to patient outcome assessment. This task is challenging for algorithms and human experts alike, with deterioration of algorithmic performance under shifts in image representations. Considerable covariate shifts occur when assessment is performed on different tumor types, images are acquired using different digitization devices, or specimens are produced in different laboratories. This observation motivated the inception of the 2022 challenge on MItosis Domain Generalization (MIDOG 2022). The challenge provided annotated histologic tumor images from six different domains and evaluated the algorithmic approaches for mitotic figure detection provided by nine challenge participants on ten independent domains. Ground truth for mitotic figure detection was established in two ways: a three-expert majority vote and an independent, immunohistochemistry-assisted set of labels. This work represents an overview of the challenge tasks, the algorithmic strategies employed by the participants, and potential factors contributing to their success. With an F1 score of 0.764 for the top-performing team, we summarize that domain generalization across various tumor domains is possible with today's deep learning-based recognition pipelines. However, we also found that domain characteristics not present in the training set (feline as new species, spindle cell shape as new morphology and a new scanner) led to small but significant decreases in performance. When assessed against the immunohistochemistry-assisted reference standard, all methods resulted in reduced recall scores, with only minor changes in the order of participants in the ranking.Copyright © 2024 The Author(s). Published by Elsevier B.V. All rights reserved.

基金：
M.A. and J.A. acknowledge support from the Bavarian Research Institute for Digital Transformation (project ReGInA), Germany. M.A. and R.K. acknowledge support by the German Research Foundation, Germany (project number 520330054). K.B. and F.W. received funding by the German Research Foundation (DFG) project 460333672 CRC1540 EBM, Germany. K.B. further acknowledges support by d.hip campus – bavarian aim in form of a faculty endowment. C.A.B. acknowledges funding by the Austrian Science Fund, Austria (FWF, project number: I 6555).
语种：
外文
PubmedID：
中科院(CAS)分区：
出版当年[2023]版：
大类 | 1 区 医学
小类 | 1 区 计算机：人工智能 1 区 计算机：跨学科应用 1 区 工程：生物医学 1 区 核医学
最新[2023]版：
大类 | 1 区 医学
小类 | 1 区 计算机：人工智能 1 区 计算机：跨学科应用 1 区 工程：生物医学 1 区 核医学
第一作者：
第一作者机构： [1]Technische Hochschule Ingolstadt, Ingolstadt, Germany
通讯作者：
推荐引用方式(GB/T 7714)：
Marc Aubreville,Nikolas Stathonikos,Taryn A. Donovan,et al.Domain generalization across tumor types, laboratories, and species - Insights from the 2022 edition of the Mitosis Domain Generalization Challenge[J].Medical Image Analysis.2024,94:103155.doi:10.1016/j.media.2024.103155.
APA：
Marc Aubreville,Nikolas Stathonikos,Taryn A. Donovan,Robert Klopfleisch,Jonas Ammeling...&Christof A. Bertram.(2024).Domain generalization across tumor types, laboratories, and species - Insights from the 2022 edition of the Mitosis Domain Generalization Challenge.Medical Image Analysis,94,
MLA：
Marc Aubreville,et al."Domain generalization across tumor types, laboratories, and species - Insights from the 2022 edition of the Mitosis Domain Generalization Challenge".Medical Image Analysis 94.(2024):103155

相关文献

[1]Mitosis domain generalization in histopathology images - The MIDOG challenge [2]CoNIC Challenge: Pushing the frontiers of nuclear detection, segmentation, classification and counting [3]CDDSA: Contrastive domain disentanglement and style augmentation for generalizable medical image segmentation [4]A Deep Learning-Based Radiomics Model for Prediction of Survival in Glioblastoma Multiforme [5]Predicting cutaneous malignant melanoma patients' survival using deep learning: a retrospective cohort study [6]Efficient application of deep learning-based elective lymph node regions delineation for pelvic malignancies [7]Deep learning based clinical target volumes contouring for prostate cancer: Easy and efficient application [8]A novel deep learning segmentation model for organoid-based drug screening [9]Domain composition and attention network trained with synthesized unlabeled images for generalizable medical image segmentation [10]Using deep learning to classify pediatric posttraumatic stress disorder at the individual level.

资源点击量：43377 今日访问量：0 总访问量：3120 更新日期：2024-09-01 建议使用谷歌、火狐浏览器 常见问题

版权所有©2020 四川省肿瘤医院 技术支持：重庆聚合科技有限公司 地址：成都市人民南路四段55号