Self-supervision enhances instance-based multiple instance learning methods in digital pathology: a benchmark study.

IF 1.9 Q3 RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING
Journal of Medical Imaging Pub Date : 2025-11-01 Epub Date: 2025-06-03 DOI:10.1117/1.JMI.12.6.061404
Ali Mammadov, Loïc Le Folgoc, Julien Adam, Anne Buronfosse, Gilles Hayem, Guillaume Hocquet, Pietro Gori
{"title":"Self-supervision enhances instance-based multiple instance learning methods in digital pathology: a benchmark study.","authors":"Ali Mammadov, Loïc Le Folgoc, Julien Adam, Anne Buronfosse, Gilles Hayem, Guillaume Hocquet, Pietro Gori","doi":"10.1117/1.JMI.12.6.061404","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Multiple instance learning (MIL) has emerged as the best solution for whole slide image (WSI) classification. It consists of dividing each slide into patches, which are treated as a bag of instances labeled with a global label. MIL includes two main approaches: instance-based and embedding-based. In the former, each patch is classified independently, and then, the patch scores are aggregated to predict the bag label. In the latter, bag classification is performed after aggregating patch embeddings. Even if instance-based methods are naturally more interpretable, embedding-based MILs have usually been preferred in the past due to their robustness to poor feature extractors. Recently, the quality of feature embeddings has drastically increased using self-supervised learning (SSL). Nevertheless, many authors continue to endorse the superiority of embedding-based MIL.</p><p><strong>Approach: </strong>We conduct 710 experiments across 4 datasets, comparing 10 MIL strategies, 6 self-supervised methods with 4 backbones, 4 foundation models, and various pathology-adapted techniques. Furthermore, we introduce 4 instance-based MIL methods, never used before in the pathology domain.</p><p><strong>Results: </strong>We show that with a good SSL feature extractor, simple instance-based MILs, with very few parameters, obtain similar or better performance than complex, state-of-the-art (SOTA) embedding-based MIL methods, setting new SOTA results on the BRACS and Camelyon16 datasets.</p><p><strong>Conclusion: </strong>As simple instance-based MIL methods are naturally more interpretable and explainable to clinicians, our results suggest that more effort should be put into well-adapted SSL methods for WSI rather than into complex embedding-based MIL methods.</p>","PeriodicalId":47707,"journal":{"name":"Journal of Medical Imaging","volume":"12 6","pages":"061404"},"PeriodicalIF":1.9000,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12134610/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Medical Imaging","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1117/1.JMI.12.6.061404","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/6/3 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING","Score":null,"Total":0}
引用次数: 0

Abstract

Purpose: Multiple instance learning (MIL) has emerged as the best solution for whole slide image (WSI) classification. It consists of dividing each slide into patches, which are treated as a bag of instances labeled with a global label. MIL includes two main approaches: instance-based and embedding-based. In the former, each patch is classified independently, and then, the patch scores are aggregated to predict the bag label. In the latter, bag classification is performed after aggregating patch embeddings. Even if instance-based methods are naturally more interpretable, embedding-based MILs have usually been preferred in the past due to their robustness to poor feature extractors. Recently, the quality of feature embeddings has drastically increased using self-supervised learning (SSL). Nevertheless, many authors continue to endorse the superiority of embedding-based MIL.

Approach: We conduct 710 experiments across 4 datasets, comparing 10 MIL strategies, 6 self-supervised methods with 4 backbones, 4 foundation models, and various pathology-adapted techniques. Furthermore, we introduce 4 instance-based MIL methods, never used before in the pathology domain.

Results: We show that with a good SSL feature extractor, simple instance-based MILs, with very few parameters, obtain similar or better performance than complex, state-of-the-art (SOTA) embedding-based MIL methods, setting new SOTA results on the BRACS and Camelyon16 datasets.

Conclusion: As simple instance-based MIL methods are naturally more interpretable and explainable to clinicians, our results suggest that more effort should be put into well-adapted SSL methods for WSI rather than into complex embedding-based MIL methods.

自我监督增强基于实例的多实例学习方法在数字病理学:一个基准研究。
目的:多实例学习(MIL)已成为全幻灯片图像(WSI)分类的最佳方法。它包括将每张幻灯片划分为补丁,这些补丁被视为带有全局标签的实例包。MIL包括两种主要方法:基于实例的和基于嵌入的。前一种方法是对每个贴片进行独立分类,然后将贴片得分进行汇总来预测包装袋标签。在后者中,袋分类是在聚集补丁嵌入后进行的。即使基于实例的方法自然地更具可解释性,基于嵌入的mil在过去通常是首选的,因为它们对较差的特征提取器具有鲁棒性。最近,使用自监督学习(self-supervised learning, SSL),特征嵌入的质量得到了极大的提高。方法:我们在4个数据集上进行了710个实验,比较了10种MIL策略、6种具有4个主干的自监督方法、4个基础模型和各种病理适应技术。此外,我们介绍了4种基于实例的MIL方法,这些方法以前从未在病理学领域使用过。结果:我们表明,使用良好的SSL特征提取器,简单的基于实例的MIL,使用很少的参数,获得与复杂的,最先进的(SOTA)基于嵌入的MIL方法相似或更好的性能,在BRACS和Camelyon16数据集上设置新的SOTA结果。结论:由于简单的基于实例的MIL方法对临床医生来说自然更具可解释性和可解释性,我们的研究结果表明,应该更多地努力开发适合WSI的SSL方法,而不是复杂的基于嵌入的MIL方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Journal of Medical Imaging
Journal of Medical Imaging RADIOLOGY, NUCLEAR MEDICINE & MEDICAL IMAGING-
CiteScore
4.10
自引率
4.20%
发文量
0
期刊介绍: JMI covers fundamental and translational research, as well as applications, focused on medical imaging, which continue to yield physical and biomedical advancements in the early detection, diagnostics, and therapy of disease as well as in the understanding of normal. The scope of JMI includes: Imaging physics, Tomographic reconstruction algorithms (such as those in CT and MRI), Image processing and deep learning, Computer-aided diagnosis and quantitative image analysis, Visualization and modeling, Picture archiving and communications systems (PACS), Image perception and observer performance, Technology assessment, Ultrasonic imaging, Image-guided procedures, Digital pathology, Biomedical applications of biomedical imaging. JMI allows for the peer-reviewed communication and archiving of scientific developments, translational and clinical applications, reviews, and recommendations for the field.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信