Ipnet:半监督磁共振图像分割的信息补丁学习。

IF 3.2 4区 医学 Q2 ENGINEERING, BIOMEDICAL
Biomedical Engineering Letters Pub Date : 2025-05-29 eCollection Date: 2025-07-01 DOI:10.1007/s13534-025-00481-9
Guangxing Du, Rui Wu, Jinming Xu, Xiang Zeng, Shengwu Xiong
{"title":"Ipnet:半监督磁共振图像分割的信息补丁学习。","authors":"Guangxing Du, Rui Wu, Jinming Xu, Xiang Zeng, Shengwu Xiong","doi":"10.1007/s13534-025-00481-9","DOIUrl":null,"url":null,"abstract":"<p><p>Semi-supervised learning has become a favorable method for medical image segmentation due to the high cost of obtaining labeled data in the field of medical image analysis. However, existing magnetic resonance images have low contrast, the scale and shape of organs vary greatly under different slice perspectives. Although existing methods have made some progress, they still cannot handle these challenging samples well. To this end, we propose a semi-supervised magnetic resonance images segmentation method based on informative patches learning (IPNet), which focuses on the learning of challenging regions. Specifically, we design a novel informative patch scoring strategy based on prediction uncertainty and category diversity, which can accurately identify challenging regions in samples. And to ensure that the informative patch is fully learned, the patch with the lowest score in one sample is replaced with the patch with the highest score in another sample to obtain a new pair of training samples. Furthermore, we introduce global and local consistency losses to supervise the new samples, guide the model to focus on the global and local features of the informative patches. To evaluate the effectiveness of the method, we conducted experiments on three magnetic resonance image datasets (ACDC, PROMISE 12 and LA datasets). Extensive experimental results demonstrate the effectiveness and superior performance of the proposed method.</p>","PeriodicalId":46898,"journal":{"name":"Biomedical Engineering Letters","volume":"15 4","pages":"797-807"},"PeriodicalIF":3.2000,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12229363/pdf/","citationCount":"0","resultStr":"{\"title\":\"Ipnet: informative patches learning for semi-supervised magnetic resonance image segmentation.\",\"authors\":\"Guangxing Du, Rui Wu, Jinming Xu, Xiang Zeng, Shengwu Xiong\",\"doi\":\"10.1007/s13534-025-00481-9\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Semi-supervised learning has become a favorable method for medical image segmentation due to the high cost of obtaining labeled data in the field of medical image analysis. However, existing magnetic resonance images have low contrast, the scale and shape of organs vary greatly under different slice perspectives. Although existing methods have made some progress, they still cannot handle these challenging samples well. To this end, we propose a semi-supervised magnetic resonance images segmentation method based on informative patches learning (IPNet), which focuses on the learning of challenging regions. Specifically, we design a novel informative patch scoring strategy based on prediction uncertainty and category diversity, which can accurately identify challenging regions in samples. And to ensure that the informative patch is fully learned, the patch with the lowest score in one sample is replaced with the patch with the highest score in another sample to obtain a new pair of training samples. Furthermore, we introduce global and local consistency losses to supervise the new samples, guide the model to focus on the global and local features of the informative patches. To evaluate the effectiveness of the method, we conducted experiments on three magnetic resonance image datasets (ACDC, PROMISE 12 and LA datasets). Extensive experimental results demonstrate the effectiveness and superior performance of the proposed method.</p>\",\"PeriodicalId\":46898,\"journal\":{\"name\":\"Biomedical Engineering Letters\",\"volume\":\"15 4\",\"pages\":\"797-807\"},\"PeriodicalIF\":3.2000,\"publicationDate\":\"2025-05-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12229363/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Biomedical Engineering Letters\",\"FirstCategoryId\":\"5\",\"ListUrlMain\":\"https://doi.org/10.1007/s13534-025-00481-9\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/7/1 0:00:00\",\"PubModel\":\"eCollection\",\"JCR\":\"Q2\",\"JCRName\":\"ENGINEERING, BIOMEDICAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Biomedical Engineering Letters","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1007/s13534-025-00481-9","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/7/1 0:00:00","PubModel":"eCollection","JCR":"Q2","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 0

摘要

在医学图像分析领域,由于获取标记数据的成本较高,半监督学习已成为一种较好的医学图像分割方法。然而,现有的磁共振图像对比度低,在不同的切片角度下,器官的尺度和形状差异很大。虽然现有的方法已经取得了一些进展,但它们仍然不能很好地处理这些具有挑战性的样品。为此,我们提出了一种基于信息补丁学习(IPNet)的半监督磁共振图像分割方法,该方法侧重于挑战区域的学习。具体而言,我们设计了一种基于预测不确定性和类别多样性的信息补丁评分策略,可以准确识别样本中的挑战区域。为了保证信息patch被充分学习,将一个样本中得分最低的patch替换为另一个样本中得分最高的patch,得到一对新的训练样本。此外,我们引入全局和局部一致性损失来监督新样本,引导模型关注信息补丁的全局和局部特征。为了评估该方法的有效性,我们在三个磁共振图像数据集(ACDC、PROMISE 12和LA数据集)上进行了实验。大量的实验结果证明了该方法的有效性和优越的性能。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Ipnet: informative patches learning for semi-supervised magnetic resonance image segmentation.

Semi-supervised learning has become a favorable method for medical image segmentation due to the high cost of obtaining labeled data in the field of medical image analysis. However, existing magnetic resonance images have low contrast, the scale and shape of organs vary greatly under different slice perspectives. Although existing methods have made some progress, they still cannot handle these challenging samples well. To this end, we propose a semi-supervised magnetic resonance images segmentation method based on informative patches learning (IPNet), which focuses on the learning of challenging regions. Specifically, we design a novel informative patch scoring strategy based on prediction uncertainty and category diversity, which can accurately identify challenging regions in samples. And to ensure that the informative patch is fully learned, the patch with the lowest score in one sample is replaced with the patch with the highest score in another sample to obtain a new pair of training samples. Furthermore, we introduce global and local consistency losses to supervise the new samples, guide the model to focus on the global and local features of the informative patches. To evaluate the effectiveness of the method, we conducted experiments on three magnetic resonance image datasets (ACDC, PROMISE 12 and LA datasets). Extensive experimental results demonstrate the effectiveness and superior performance of the proposed method.

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
Biomedical Engineering Letters
Biomedical Engineering Letters ENGINEERING, BIOMEDICAL-
CiteScore
6.80
自引率
0.00%
发文量
34
期刊介绍: Biomedical Engineering Letters (BMEL) aims to present the innovative experimental science and technological development in the biomedical field as well as clinical application of new development. The article must contain original biomedical engineering content, defined as development, theoretical analysis, and evaluation/validation of a new technique. BMEL publishes the following types of papers: original articles, review articles, editorials, and letters to the editor. All the papers are reviewed in single-blind fashion.
×
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