Whole slide images (WSIs) are gigapixel-scale digital images of H\&E-stained tissue samples widely used in pathology. The substantial size and complexity of WSIs pose unique analytical challenges. Multiple Instance Learning (MIL) has emerged as a powerful approach for addressing these challenges, particularly in cancer classification and detection. This survey provides a comprehensive overview of the challenges and methodologies associated with applying MIL to WSI analysis, including attention mechanisms, pseudo-labeling, transformers, pooling functions, and graph neural networks. Additionally, it explores the potential of MIL in discovering cancer cell morphology, constructing interpretable machine learning models, and quantifying cancer grading. By summarizing the current challenges, methodologies, and potential applications of MIL in WSI analysis, this survey aims to inform researchers about the state of the field and inspire future research directions.
DPsurv: Dual-Prototype Evidential Fusion for Uncertainty-Aware and Interpretable Whole-Slide Image Survival Prediction
National University of Singapore;National University of Singapore Guangzhou Research Translation;Innovation Institute;Imperial College London;Peking Union Medical College Hospital, Chinese Academy of Medical Sciences \& Peking Union Medical College;Hunan University;Institute of High Performance Computing, Agency for Science, Technology;Research (A STAR)
多实例学习(MIL)是用于整个切片图像分类的优选方法。然而,大多数MIL方法不利用从整个幻灯片图像中提取的瓦片的相互依赖性,这可以为分类提供有价值的线索。本文提出了一种新的MIL方法,利用瓷砖的空间关系分类整个幻灯片图像。为此,从瓦片嵌入构建稀疏映射,然后通过稀疏输入CNN进行分类。它在涉及10000张全切片图像的癌症亚型分类方面获得了超过流行的MIL方法的最先进性能。我们的研究结果表明,所提出的方法可以(i)提高表示学习的实例和(ii)利用上下文的实例嵌入,以提高分类性能。这个工作的代码是开源的{github censored for review}。
Comparison of Different Methods for Tissue Segmentation in Histopathological Whole-Slide Images
Radboud University Medical Center(拉德堡德大学医学中心)
组织分割是数字病理学中高效准确诊断的重要前提。然而,众所周知,例如由于组织类型,或者由于弱染色,全载玻片扫描仪可能无法检测所有组织区域,因为它们的组织检测算法不够鲁棒。在本文中,我们介绍了两种不同的卷积神经网络架构的整个幻灯片图像分割,准确地识别组织切片。我们还比较了算法的出版的传统方法。我们从三个实验室收集了54张不同染色和组织类型的全载玻片图像,以验证我们的算法。我们发现,虽然这两种方法没有显着差异,但它们优于传统方法(Jaccard指数为0.937和0.929 vs. 0.870,p <0.01)。
Stony Brook University(石溪大学);Argonne National Laboratory(阿贡国家实验室);University of Chicago(芝加哥大学);The University of Utah(犹他大学);The University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校)
HistomicsML2.0: Fast interactive machine learning for whole slide imaging data
Weisberg Division of Computer Science, Marshall University(马歇尔大学魏斯伯格计算机科学分部);Kitware, Inc.(Kitware公司);Emory University School of Medicine(埃默里大学医学院);National Cancer Institute(美国国家癌症研究所);Cairo University(开罗大学);Northwestern University Feinberg School of Medicine(西北大学费恩伯格医学院)
Overcoming the limitations of patch-based learning to detect cancer in whole slide images
University of Toronto(多伦多大学);University of British Columbia(不列颠哥伦比亚大学);Sunnybrook Health Science Center(森尼布鲁克健康科学中心);Sunnybrook Research Institute(森尼布鲁克研究所)
Benchmarking Pathology Feature Extractors for Whole Slide Image Classification
University of St Andrews(圣安德鲁斯大学);Else Kröner Fresenius Center for Digital Health(埃尔瑟·克勒纳·弗雷森纽斯数字健康中心);TUD Dresden University of Technology(德累斯顿工业大学);University Hospital Aachen(亚琛大学医院);University Hospital Heidelberg(海德堡大学医院);University Hospital Dresden(德累斯顿大学医院)
Restoration of marker occluded hematoxylin and eosin stained whole slide histology images using generative adversarial networks
Image Data Analytics, Informatics IT, Merck \& Co., Inc., West Point, PA 19486, USA Image Data Analytics, Informatics IT, Merck \& Co., Inc., Rahway, NJ 07065, USA
SlideGCD: Slide-based Graph Collaborative Training with Knowledge Distillation for Whole Slide Image Classification
School of Computer Science and Information Engineering, Hefei University of Technology(合肥工业大学计算机与信息工程学院);School of Software, Hefei University of Technology(合肥工业大学软件学院);Image Processing Center, School of Astronautics, Beihang University(北京航空航天大学宇航学院图像处理中心);Beijing Advanced Innovation Center on Biomedical Engineering(北京生物医学工程高精尖创新中心)
University of Liverpool(利物浦大学);Cixi Institute of Biomedical Engineering, Chinese Academy of Sciences(中国科学院慈溪生物医学工程研究所);China Science IntelliCloud Technology Co., Ltd(中科智云科技有限公司);Remark AI UK Limited(Remark AI英国有限公司)
Imaging Informatics Division(成像信息学部门);Bioinformatics Institute(生物信息学研究所);A*STAR(新加坡科技研究局);Lucence Diagnostics(Lucence诊断公司);Institute of Bioengineering and Nanotechnology(生物工程与纳米技术研究所);Singapore General Hospital(新加坡综合医院);National Cancer Centre Singapore(新加坡国家癌症中心);Institute of Molecular and Cell Biology(分子与细胞生物学研究所);National University of Singapore(新加坡国立大学);Singapore Eye Research Institute(新加坡眼科研究所);Image and Pervasive Access Lab(图像与普适访问实验室)
Interactive Whole Slide Images for RL-based Tumour Segmentation
Université Côte d’Azur(蔚蓝海岸大学);Inria(法国国家信息与自动化研究所);CNRS(法国国家科学研究中心);INSERM(法国国家健康与医学研究院);IBV(生物信息研究所);Politecnico di Torino(都灵理工大学);Bichat Hospital(比沙医院);Assistance Publique–Hôpitaux de Paris(巴黎公共医疗集团)
Federated Learning for Computational Pathology on Gigapixel Whole Slide Images
Brigham and Women’s Hospital(布里格姆妇女医院);Harvard Medical School(哈佛医学院);Broad Institute of Harvard and MIT(哈佛-麻省理工学院布罗德研究所);Department of Pathology(病理系);Northwell Health(诺斯韦尔健康);Department of Biomedical Informatics(生物医学信息学系);Dana-Farber Cancer Institute(达纳-法伯癌症研究所);Cancer Program(癌症项目);Cancer Data Science(癌症数据科学)
Paired Uterine Whole-Slide Images and Pathology Reports for Multimodal Computational Pathology
Institute of Pathology, Technical University of Munich(慕尼黑工业大学病理研究所);Computer Aided Medical Procedures (CAMP), Technical University of Munich(慕尼黑工业大学计算机辅助医疗程序(CAMP));Munich Center for Machine Learning (MCML)(慕尼黑机器学习中心);Department of Human Pathology, Juntendo University Graduate School of Medicine(顺天堂大学医学研究生院人体病理学部);The Hong Kong University of Science and Technology(香港科技大学);Munich Data Science Institute (MDSI)(慕尼黑数据科学研究所)
RMDL: Recalibrated multi-instance deep learning for whole slide gastric image classification
The Chinese University of Hong Kong(香港中文大学);the Sixth Affiliated Hospital of Sun Yat-sen University(中山大学附属第六医院);Imsight Medical Technology Co., Ltd.(艾迈医疗科技有限公司)
Whole-Slide Image Analysis under Realistic Few-Shot Annotation Protocols
Institute of Information and Communication Technologies, Electronics, and Applied Mathematics (ICTEAM), Université catholique de Louvain (UCLouvain)(天主教鲁汶大学信息与通信技术、电子与应用数学研究所)
针对现实 whole-slide 图像场景的小样本标注问题,提出 SlideCRF 模型,结合空间与生物线索优化预测,在四个数据集上宏观 F1 较现有转导方法更优,零样本预测提升显著。