Label Co-Occurrence Learning With Graph Convolutional Networks for Multi-Label Chest X-Ray Image Classification

计算机科学 人工智能 卷积神经网络 模式识别(心理学) 多标签分类 上下文图像分类 图形 图像(数学) 理论计算机科学
作者
Bingzhi Chen,Jinxing Li,Guangming Lu,Hongbing Yu,David Zhang
出处
期刊:IEEE Journal of Biomedical and Health Informatics [Institute of Electrical and Electronics Engineers]
卷期号:24 (8): 2292-2302 被引量:118
标识
DOI:10.1109/jbhi.2020.2967084
摘要

Existing multi-label medical image learning tasks generally contain rich relationship information among pathologies such as label co-occurrence and interdependency, which is of great importance for assisting in clinical diagnosis and can be represented as the graph-structured data. However, most state-of-the-art works only focus on regression from the input to the binary labels, failing to make full use of such valuable graph-structured information due to the complexity of graph data. In this paper, we propose a novel label co-occurrence learning framework based on Graph Convolution Networks (GCNs) to explicitly explore the dependencies between pathologies for the multi-label chest X-ray (CXR) image classification task, which we term the "CheXGCN". Specifically, the proposed CheXGCN consists of two modules, i.e., the image feature embedding (IFE) module and label co-occurrence learning (LCL) module. Thanks to the LCL model, the relationship between pathologies is generalized into a set of classifier scores by introducing the word embedding of pathologies and multi-layer graph information propagation. During end-to-end training, it can be flexibly integrated into the IFE module and then adaptively recalibrate multi-label outputs with these scores. Extensive experiments on the ChestX-Ray14 and CheXpert datasets have demonstrated the effectiveness of CheXGCN as compared with the state-of-the-art baselines.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
1秒前
酷波er应助蟹老板采纳,获得10
2秒前
韭黄完成签到,获得积分20
4秒前
呆萌芙蓉完成签到 ,获得积分0
5秒前
卓垚完成签到,获得积分10
6秒前
调皮的蓝天完成签到 ,获得积分10
7秒前
tongkaibing完成签到,获得积分10
8秒前
13秒前
科研怪人完成签到 ,获得积分10
14秒前
善善完成签到 ,获得积分10
15秒前
忧伤的八宝粥完成签到,获得积分0
16秒前
月儿完成签到 ,获得积分10
16秒前
韭菜盒子完成签到,获得积分20
17秒前
青衫完成签到 ,获得积分10
19秒前
kevin完成签到,获得积分10
21秒前
susu发布了新的文献求助20
23秒前
lqz07完成签到,获得积分10
25秒前
SciGPT应助蟹老板采纳,获得10
27秒前
爱科学完成签到 ,获得积分10
29秒前
笨笨凡松完成签到,获得积分10
30秒前
lydy1993完成签到,获得积分10
32秒前
时米米米完成签到,获得积分10
32秒前
每天都很忙完成签到 ,获得积分10
33秒前
orixero应助susu采纳,获得10
34秒前
Jeffrey完成签到,获得积分10
35秒前
等待的代容完成签到,获得积分10
36秒前
37秒前
灵巧胜完成签到 ,获得积分10
37秒前
量子星尘发布了新的文献求助10
41秒前
Tin完成签到,获得积分10
41秒前
47秒前
张张张xxx完成签到,获得积分10
47秒前
科研通AI2S应助蟹老板采纳,获得10
48秒前
淡然一德完成签到,获得积分10
48秒前
独指蜗牛完成签到 ,获得积分10
50秒前
51秒前
51秒前
51秒前
TMOMOR应助科研通管家采纳,获得10
51秒前
彭于晏应助科研通管家采纳,获得10
51秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Picture Books with Same-sex Parented Families: Unintentional Censorship 700
ACSM’s Guidelines for Exercise Testing and Prescription, 12th edition 500
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
Indomethacinのヒトにおける経皮吸収 400
Phylogenetic study of the order Polydesmida (Myriapoda: Diplopoda) 370
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3976745
求助须知:如何正确求助?哪些是违规求助? 3520831
关于积分的说明 11204951
捐赠科研通 3257684
什么是DOI,文献DOI怎么找? 1798834
邀请新用户注册赠送积分活动 877897
科研通“疑难数据库(出版商)”最低求助积分说明 806663