Multi-graph Networks with Graph Pooling for COVID-19 Diagnosis

IF 4.9 3区 计算机科学 Q1 ENGINEERING, MULTIDISCIPLINARY
Chaosheng Tang, Wenle Xu, Junding Sun, Shuihua Wang, Yudong Zhang, Juan Manuel Górriz
{"title":"Multi-graph Networks with Graph Pooling for COVID-19 Diagnosis","authors":"Chaosheng Tang,&nbsp;Wenle Xu,&nbsp;Junding Sun,&nbsp;Shuihua Wang,&nbsp;Yudong Zhang,&nbsp;Juan Manuel Górriz","doi":"10.1007/s42235-024-00600-9","DOIUrl":null,"url":null,"abstract":"<div><p>Convolutional Neural Networks (CNNs) have shown remarkable capabilities in extracting local features from images, yet they often overlook the underlying relationships between pixels. To address this limitation, previous approaches have attempted to combine CNNs with Graph Convolutional Networks (GCNs) to capture global features. However, these approaches typically neglect the topological structure information of the graph during the global feature extraction stage. This paper proposes a novel end-to-end hybrid architecture called the Multi-Graph Pooling Network (MGPN), which is designed explicitly for chest X-ray image classification. Our approach sequentially combines CNNs and GCNs, enabling the learning of both local and global features from individual images. Recognizing that different nodes contribute differently to the final graph representation, we introduce an NI-GTP module to enhance the extraction of ultimate global features. Additionally, we introduce a G-LFF module to fuse the local and global features effectively.</p></div>","PeriodicalId":614,"journal":{"name":"Journal of Bionic Engineering","volume":"21 6","pages":"3179 - 3200"},"PeriodicalIF":4.9000,"publicationDate":"2024-11-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Bionic Engineering","FirstCategoryId":"94","ListUrlMain":"https://link.springer.com/article/10.1007/s42235-024-00600-9","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

Abstract

Convolutional Neural Networks (CNNs) have shown remarkable capabilities in extracting local features from images, yet they often overlook the underlying relationships between pixels. To address this limitation, previous approaches have attempted to combine CNNs with Graph Convolutional Networks (GCNs) to capture global features. However, these approaches typically neglect the topological structure information of the graph during the global feature extraction stage. This paper proposes a novel end-to-end hybrid architecture called the Multi-Graph Pooling Network (MGPN), which is designed explicitly for chest X-ray image classification. Our approach sequentially combines CNNs and GCNs, enabling the learning of both local and global features from individual images. Recognizing that different nodes contribute differently to the final graph representation, we introduce an NI-GTP module to enhance the extraction of ultimate global features. Additionally, we introduce a G-LFF module to fuse the local and global features effectively.

Abstract Image

用于 COVID-19 诊断的图形池化多图网络
卷积神经网络(CNN)在从图像中提取局部特征方面表现出了非凡的能力,但它们往往忽略了像素之间的潜在关系。为了解决这一局限性,以前的方法尝试将 CNN 与图卷积网络 (GCN) 结合起来,以捕捉全局特征。然而,这些方法在全局特征提取阶段通常会忽略图的拓扑结构信息。本文提出了一种新颖的端到端混合架构,称为多图池化网络(MGPN),专门用于胸部 X 光图像分类。我们的方法依次结合了 CNN 和 GCN,从而能够从单个图像中学习局部和全局特征。考虑到不同节点对最终图表示的贡献不同,我们引入了 NI-GTP 模块,以增强最终全局特征的提取。此外,我们还引入了 G-LFF 模块,以有效融合局部和全局特征。
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 Bionic Engineering
Journal of Bionic Engineering 工程技术-材料科学:生物材料
CiteScore
7.10
自引率
10.00%
发文量
162
审稿时长
10.0 months
期刊介绍: The Journal of Bionic Engineering (JBE) is a peer-reviewed journal that publishes original research papers and reviews that apply the knowledge learned from nature and biological systems to solve concrete engineering problems. The topics that JBE covers include but are not limited to: Mechanisms, kinematical mechanics and control of animal locomotion, development of mobile robots with walking (running and crawling), swimming or flying abilities inspired by animal locomotion. Structures, morphologies, composition and physical properties of natural and biomaterials; fabrication of new materials mimicking the properties and functions of natural and biomaterials. Biomedical materials, artificial organs and tissue engineering for medical applications; rehabilitation equipment and devices. Development of bioinspired computation methods and artificial intelligence for engineering applications.
×
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