GLRP: Global and local contrastive learning based on relative position for medical image segmentation on cardiac MRI

Author:

Zhao Xin1ORCID,Wang Tongming1ORCID,Chen Jingsong1ORCID,Jiang Bingrun1ORCID,Li Haotian1ORCID,Zhang Nan2ORCID,Yang Guang3456ORCID,Chai Senchun1ORCID

Affiliation:

1. School of Automation Beijing Institute of Technology Beijing China

2. Department of Radiology, Beijing Anzhen Hospital Capital Medical University Beijing China

3. National Heart and Lung Institute Imperial College London London UK

4. Bioengineering Department and Imperial‐X Imperial College London London UK

5. Cardiovascular Research Centre Royal Brompton Hospital London UK

6. School of Biomedical Engineering & Imaging Sciences King's College London London UK

Abstract

AbstractContrastive learning, as an unsupervised technique, is widely employed in image segmentation to enhance segmentation performance even when working with small labeled datasets. However, generating positive and negative data pairs for medical image segmentation poses a challenge due to the presence of similar tissues and organs across different slices in datasets. To tackle this issue, we propose a novel contrastive learning strategy that leverages the relative position differences between image slices. Additionally, we combine global and local features to address this problem effectively. In order to enhance segmentation accuracy and reduce isolated mis‐segmented regions, we employ a two‐dimensional fully connected conditional random field for iterative optimization of the segmentation results. With only 10 labeled samples, our proposed method is able to achieve average dice scores of 0.876 and 0.899 on the public and private dataset heart segmentation tasks, surpassing the PCL method's 0.801 and 0.852. Experimental results on both public and private MRI datasets demonstrate that our proposed method yields significant improvements in medical segmentation tasks with limited annotated samples, outperforming existing semi‐supervised and self‐supervised techniques.

Funder

Horizon 2020 Framework Programme

Royal Society

Publisher

Wiley

Subject

Electrical and Electronic Engineering,Computer Vision and Pattern Recognition,Software,Electronic, Optical and Magnetic Materials

Reference34 articles.

1. Automated segmentation of the left ventricle in cardiac MRI

2. Fully convolutional multi-scale residual DenseNets for cardiac segmentation and automated cardiac diagnosis using ensemble of classifiers

3. Fully convolutional networks for semantic segmentation

4. IsenseeF PetersenJ KleinA et al.Nnu‐net: self‐adapting framework for u‐net‐based medical image segmentation[J]. arXiv Preprint arXiv:1809.10486.2018.

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Self-Supervised Contrastive Learning for Automated Segmentation of Brain Tumor MRI Images in Schizophrenia;International Journal of Computational Intelligence Systems;2024-07-29

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3