Are pathologist-defined labels reproducible? Comparison of the TUPAC16 mitotic figure dataset with an alternative set of labels
Freie Universität Berlin（柏林自由大学）；Eindhoven University of Technology（埃因霍温理工大学）；Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡弗里德里希-亚历山大大学）；University Medical Center Utrecht（乌得勒支大学医学中心）
Christof A. Bertram, Mitko Veta, Christian Marzahl, Nikolas Stathonikos, Andreas Maier, Robert Klopfleisch, Marc Aubreville
In: Cardoso J. et al. (eds) Interpretable and Annotation-Efficient Learning for Medical Image Computing. IMIMIC 2020, MIL3ID 2020, LABELS 2020. Lecture Notes in Computer Science, vol 12446. Springer, Cham
Pathologist-defined labels are the gold standard for histopathological data sets, regardless of well-known limitations in consistency for some tasks. To date, some datasets on mitotic figures are available and were used for development of promising deep learning-based algorithms. In order to assess robustness of those algorithms and reproducibility of their methods it is necessary to test on several independent datasets. The influence of different labeling methods of these available datasets is currently unknown. To tackle this, we present an alternative set of labels for the images of the auxiliary mitosis dataset of the TUPAC16 challenge. Additional to manual mitotic figure screening, we used a novel, algorithm-aided labeling process, that allowed to minimize the risk of missing rare mitotic figures in the images. All potential mitotic figures were independently assessed by two pathologists. The novel, publicly available set of labels contains 1,999 mitotic figures (+28.80%) and additionally includes 10,483 labels of cells with high similarities to mitotic figures (hard examples). We found significant difference comparing F_1 scores between the original label set (0.549) and the new alternative label set (0.735) using a standard deep learning object detection architecture. The models trained on the alternative set showed higher overall confidence values, suggesting a higher overall label consistency. Findings of the present study show that pathologists-defined labels may vary significantly resulting in notable difference in the model performance. Comparison of deep learning-based algorithms between independent datasets with different labeling methods should be done with caution.
Deep learning algorithms out-perform veterinary pathologists in detecting the mitotically most active tumor region
Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡弗里德里希-亚历山大大学）；Freie Universität Berlin（柏林自由大学）；University of Bern（伯尔尼大学）；IDEXX Laboratories（艾迪克斯实验室）
How Many Annotators Do We Need? -- A Study on the Influence of Inter-Observer Variability on the Reliability of Automatic Mitotic Figure Assessment
Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡弗里德里希-亚历山大大学）；Freie Universität Berlin（柏林自由大学）；Animal Medical Center（动物医疗中心）；University of Pennsylvania（宾夕法尼亚大学）；University of Veterinary Medicine Hannover（汉诺威兽医大学）；Synlab’s VPG Histology（Synlab VPG组织学部门）；Michigan State University（密歇根州立大学）；Idorsia Pharmaceuticals Ltd（Idorsia制药有限公司）；International Zoo Veterinary Group（国际动物园兽医集团）；University of Bern（伯尔尼大学）；IDEXX Vet Med Labor GmbH（爱德士兽医实验室有限责任公司）
Field Of Interest Proposal for Augmented Mitotic Cell Count: Comparison of two Convolutional Networks
Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡弗里德里希-亚历山大大学）；Freie Universität Berlin（柏林自由大学）
大多数人类肿瘤分级系统以及兽医组织病理学分级系统都是基于组织学切片特定参考区域中有丝分裂像的绝对计数。由于在诊断环境中，用于鉴定的时间是有限的，因此病理学家通常几乎任意地选择假定具有最高有丝分裂活性的感兴趣的某个领域。然而，由于有丝分裂像在载玻片上通常非常稀疏，并且通常具有斑片状分布，这造成了已知能够影响肿瘤定位的采样问题。另一方面，核分裂像的自动检测还不能被认为是足够可靠的临床应用。为了帮助人类专家的工作，同时减少肿瘤分级的差异，评估整个载玻片图像（WSI）的最高有丝分裂活性并将其用作人类计数的参考区域是有益的。对于这项任务，我们比较了两种基于卷积神经网络（CNN）的感兴趣区域建议方法。对于这两种方法，CNN执行WSI的分割以评估有丝分裂活性。第一种方法以原始图像分辨率执行分割，而第二种方法以显著降低的分辨率执行分割操作，从而降低处理复杂度。我们使用32个完全注释的犬肥大细胞肿瘤全切片图像的数据集来评估该方法，其中22个用于网络训练，10个用于测试。我们的研究结果表明，虽然基于精细分辨率网络的方法与真实有丝分裂活性的总体相关性相当高（0.94 vs. 0.83），但感兴趣的领域选择仅略好。这两种方法都提出了在各自载玻片的上四分位数中包含有丝分裂计数的感兴趣区域。
A novel dataset and a two-stage mitosis nuclei detection method based on hybrid anchor branch
Guilin University of Electronic Technology（桂林电子科技大学）；Guangxi Key Laboratory of Image and Graphic Intelligent Processing（广西图像图形智能处理重点实验室）；Ganzhou Municipal Hospital（赣州市立医院）
Benchmarking Deep Learning and Vision Foundation Models for Atypical vs. Normal Mitosis Classification with Cross-Dataset Evaluation
Flensburg University of Applied Sciences（弗劳恩霍夫应用科技大学）；University of Veterinary Medicine, Vienna（维也纳兽医大学）；The Schwarzman Animal Medical Center（施瓦茨曼动物医疗中心）；Diffusely（扩散地）；Freie Universität Berlin（柏林自由大学）；Technische Hochschule Ingolstadt（因戈尔施塔特技术大学）；Medical University of Vienna（维也纳医学大学）；Julius-Maximilians-Universität Würzburg（维尔茨堡约瑟夫-马克斯姆-大学）
SWAN -- Enabling Fast and Mobile Histopathology Image Annotation through Swipeable Interfaces
Flensburg University of Applied Sciences（弗劳恩霍夫应用科技大学）；University of Veterinary Medicine（兽医大学）；Schwarzman Animal Medical Center（施瓦茨曼动物医疗中心）；Medical University of Vienna（维也纳医学大学）；Technische Hochschule Ingolstadt（英戈尔施塔特技术大学）
Mitosis domain generalization in histopathology images -- The MIDOG challenge
Technische Hochschule Ingolstadt（因戈尔施塔特应用技术大学）；UMC Utrecht（乌得勒支大学医学中心）；Freie Universität Berlin（柏林自由大学）；Radboud UMC Nijmegen（奈梅亨拉德堡德大学医学中心）；Friedrich-Alexander-Universität Erlangen-Nürnberg（埃尔朗根-纽伦堡弗里德里希-亚历山大大学）；Schwarzman Animal Medical Center（施瓦茨曼动物医学中心）；University of Leeds（利兹大学）；Korea Advanced Institute of Science and Technology（韩国科学技术院）
Teacher-Student Model for Detecting and Classifying Mitosis in the MIDOG 2025 Challenge
Image Analysis in Medicine Lab (IAMLAB)（医学图像分析实验室）；Electrical, Computer and Biomedical Engineering（电气、计算机与生物医学工程）；Toronto Metropolitan University（多伦多 Metropolitan 大学）；Princess Margaret Cancer Centre（公主玛格丽特癌症中心）；University Health Network（大学健康网络）；Department of Laboratory Medicine and Pathobiology（实验室医学与病理学系）；University of Toronto（多伦多大学）；Vector Institute of Artificial Intelligence（人工智能向量研究所）
Assessment of algorithms for mitosis detection in breast cancer histopathology images
Image Sciences Institute, University Medical Center Utrecht（乌得勒支大学医学中心影像科学研究所）；Pathology Department, University Medical Center Utrecht（乌得勒支大学医学中心病理系）；BME Department, Case Western Reserve University（凯斯西储大学生物医学工程系）；MindLab, National University of Colombia（哥伦比亚国立大学思维实验室）；Department of Applied Mathematics and Computer Science, Technical University of Denmark（丹麦技术大学应用数学与计算机科学系）；IDSIA, USI-SUPSI（瑞士应用科技大学智能系统与数据分析研究所）；Department of Computer Engineering, Işık University（伊希克大学计算机工程系）；MINES ParisTech, PSL-Research University, CBIO-Centre for Computational Biology（巴黎高等矿业学院PSL研究大学生物计算中心）；Institut Curie（居里研究所）；INSERM U900（法国国家健康与医学研究院U900）；Graduate Institute of Biomedical Engineering, National Taiwan University of Science and Technology（台湾科技大学生物医学工程研究所）；Corporate R&D Center, Panasonic Healthcare Co., Ltd.（松下医疗保健公司企业研发中心）；School of Computing, Engineering and Physical Sciences, University of Central Lancashire（中央兰开夏大学计算、工程与物理科学学院）；I3S, UMR 7271 UNS-CNRS, Nice Sophia-Antipolis University（尼斯索菲亚安蒂波利斯大学I3S联合研究实验室）；CVSSP, University of Surrey（萨里大学视觉、声音和感知安全研究中心）；Department of Computer Science and Sheffield institute for Translational Medicine, University of Sheffield（谢菲尔德大学计算机科学与谢菲尔德转化医学研究所）；Department of Computer Science, University of Warwick（华威大学计算机系）；Department of Computer Science & Engineering, College of Engineering Qatar University（卡塔尔大学工程学院计算机科学与工程系）；The Michael Letcher Department of Cellular Pathology, Princess Alexandra Hospital NHS Trust（公主亚历山医院NHS信托迈克尔·莱切尔细胞病理学系）
