弱监督位置对比学习在肝硬化分类中的应用

Emma Sarfati, Alexandre Bône, Marc-Michel Roh'e, P. Gori, I. Bloch
{"title":"弱监督位置对比学习在肝硬化分类中的应用","authors":"Emma Sarfati, Alexandre Bône, Marc-Michel Roh'e, P. Gori, I. Bloch","doi":"10.48550/arXiv.2307.04617","DOIUrl":null,"url":null,"abstract":"Large medical imaging datasets can be cheaply and quickly annotated with low-confidence, weak labels (e.g., radiological scores). Access to high-confidence labels, such as histology-based diagnoses, is rare and costly. Pretraining strategies, like contrastive learning (CL) methods, can leverage unlabeled or weakly-annotated datasets. These methods typically require large batch sizes, which poses a difficulty in the case of large 3D images at full resolution, due to limited GPU memory. Nevertheless, volumetric positional information about the spatial context of each 2D slice can be very important for some medical applications. In this work, we propose an efficient weakly-supervised positional (WSP) contrastive learning strategy where we integrate both the spatial context of each 2D slice and a weak label via a generic kernel-based loss function. We illustrate our method on cirrhosis prediction using a large volume of weakly-labeled images, namely radiological low-confidence annotations, and small strongly-labeled (i.e., high-confidence) datasets. The proposed model improves the classification AUC by 5% with respect to a baseline model on our internal dataset, and by 26% on the public LIHC dataset from the Cancer Genome Atlas. The code is available at: https://github.com/Guerbet-AI/wsp-contrastive.","PeriodicalId":18289,"journal":{"name":"Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention","volume":"47 1","pages":"227-237"},"PeriodicalIF":0.0000,"publicationDate":"2023-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Weakly-supervised positional contrastive learning: application to cirrhosis classification\",\"authors\":\"Emma Sarfati, Alexandre Bône, Marc-Michel Roh'e, P. Gori, I. Bloch\",\"doi\":\"10.48550/arXiv.2307.04617\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Large medical imaging datasets can be cheaply and quickly annotated with low-confidence, weak labels (e.g., radiological scores). Access to high-confidence labels, such as histology-based diagnoses, is rare and costly. Pretraining strategies, like contrastive learning (CL) methods, can leverage unlabeled or weakly-annotated datasets. These methods typically require large batch sizes, which poses a difficulty in the case of large 3D images at full resolution, due to limited GPU memory. Nevertheless, volumetric positional information about the spatial context of each 2D slice can be very important for some medical applications. In this work, we propose an efficient weakly-supervised positional (WSP) contrastive learning strategy where we integrate both the spatial context of each 2D slice and a weak label via a generic kernel-based loss function. We illustrate our method on cirrhosis prediction using a large volume of weakly-labeled images, namely radiological low-confidence annotations, and small strongly-labeled (i.e., high-confidence) datasets. The proposed model improves the classification AUC by 5% with respect to a baseline model on our internal dataset, and by 26% on the public LIHC dataset from the Cancer Genome Atlas. The code is available at: https://github.com/Guerbet-AI/wsp-contrastive.\",\"PeriodicalId\":18289,\"journal\":{\"name\":\"Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention\",\"volume\":\"47 1\",\"pages\":\"227-237\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-07-10\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.48550/arXiv.2307.04617\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.48550/arXiv.2307.04617","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

大型医学成像数据集可以用低置信度、弱标签(例如放射学评分)廉价、快速地进行注释。获得高可信度的标签,如基于组织学的诊断,是罕见和昂贵的。预训练策略,如对比学习(CL)方法,可以利用未标记或弱注释的数据集。这些方法通常需要大量的批量处理,由于GPU内存有限,这在全分辨率的大型3D图像的情况下带来了困难。然而,关于每个二维切片的空间背景的体积位置信息对于某些医学应用可能非常重要。在这项工作中,我们提出了一种有效的弱监督位置(WSP)对比学习策略,我们通过通用的基于核的损失函数整合每个2D切片的空间上下文和弱标签。我们使用大量弱标记图像(即放射学低置信度注释)和小的强标记(即高置信度)数据集来说明我们的肝硬化预测方法。与我们内部数据集的基线模型相比,所提出的模型将分类AUC提高了5%,在来自癌症基因组图谱的公共LIHC数据集上提高了26%。代码可从https://github.com/Guerbet-AI/wsp-contrastive获得。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Weakly-supervised positional contrastive learning: application to cirrhosis classification
Large medical imaging datasets can be cheaply and quickly annotated with low-confidence, weak labels (e.g., radiological scores). Access to high-confidence labels, such as histology-based diagnoses, is rare and costly. Pretraining strategies, like contrastive learning (CL) methods, can leverage unlabeled or weakly-annotated datasets. These methods typically require large batch sizes, which poses a difficulty in the case of large 3D images at full resolution, due to limited GPU memory. Nevertheless, volumetric positional information about the spatial context of each 2D slice can be very important for some medical applications. In this work, we propose an efficient weakly-supervised positional (WSP) contrastive learning strategy where we integrate both the spatial context of each 2D slice and a weak label via a generic kernel-based loss function. We illustrate our method on cirrhosis prediction using a large volume of weakly-labeled images, namely radiological low-confidence annotations, and small strongly-labeled (i.e., high-confidence) datasets. The proposed model improves the classification AUC by 5% with respect to a baseline model on our internal dataset, and by 26% on the public LIHC dataset from the Cancer Genome Atlas. The code is available at: https://github.com/Guerbet-AI/wsp-contrastive.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
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:481959085
Book学术官方微信