Detection of Coronary Artery Disease Based on Clinical Phonocardiogram and Multiscale Attention Convolutional Compression Network

作者全名："Yin, Chongbo; Zheng, Yineng; Ding, Xiaorong; Shi, Yan; Qin, Jian; Guo, Xingming"

作者地址："[Yin, Chongbo; Guo, Xingming] Chongqing Univ, Coll Bioengn, Key Lab Biorheol Sci & Technol, Minist Educ, Chongqing 400044, Peoples R China; [Zheng, Yineng] Chongqing Med Univ, Affiliated Hosp 1, Dept Radiol, Chongqing 400016, Peoples R China; [Ding, Xiaorong] Univ Elect Sci & Technol China, Sch Life Sci & Technol, Chengdu 610056, Peoples R China; [Shi, Yan] North East Elect Power Univ, Sch Automat Engn, Jilin 132012, Peoples R China; [Qin, Jian] Chongqing Med Univ, Affiliated Hosp 1, Dept Cardiol, Chongqing 400016, Peoples R China"

通信作者："Guo, XM (通讯作者)，Chongqing Univ, Coll Bioengn, Key Lab Biorheol Sci & Technol, Minist Educ, Chongqing 400044, Peoples R China.; Qin, J (通讯作者)，Chongqing Med Univ, Affiliated Hosp 1, Dept Cardiol, Chongqing 400016, Peoples R China."

来源：IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS

ESI学科分类：COMPUTER SCIENCE

WOS号：WOS:001180907300004

JCR分区：Q1

影响因子：7.7

年份：2024

卷号：28

期号：3

开始页：1353

结束页：1362

文献类型：Article

关键词：Feature extraction; Phonocardiography; Convolution; Heart; Data mining; Arteries; Solid modeling; Heart sounds; coronary artery disease; convolution attention neural network; attention module

摘要："Heart sound is an important physiological signal that contains rich pathological information related with coronary stenosis. Thus, some machine learning methods are developed to detect coronary artery disease (CAD) based on phonocardiogram (PCG). However, current methods lack sufficient clinical dataset and fail to achieve efficient feature utilization. Besides, the methods require complex processing steps including empirical feature extraction and classifier design. To achieve efficient CAD detection, we propose the multiscale attention convolutional compression network (MACCN) based on clinical PCG dataset. Firstly, PCG dataset including 102 CAD subjects and 82 non-CAD subjects was established. Then, a multiscale convolution structure was developed to catch comprehensive heart sound features and a channel attention module was developed to enhance key features in multiscale attention convolutional block (MACB). Finally, a separate downsampling block was proposed to reduce feature losses. MACCN combining the blocks can automatically extract features without empirical and manual feature selection. It obtains good classification results with accuracy 93.43%, sensitivity 93.44%, precision 93.48%, and F1 score 93.42%. The study implies that MACCN performs effective PCG feature mining aiming for CAD detection. Further, it integrates feature extraction and classification and provides a simplified PCG processing case."

基金机构：National Natural Science Foundation of China

基金资助正文：No Statement Available