电子学报 ›› 2021, Vol. 49 ›› Issue (6): 1217-1223.DOI: 10.12263/DZXB.20200907

• 学术论文 • 上一篇    下一篇

基于EWT和结构特征提取的心电信号R波识别算法

林金朝1,3, 李必禄1,3, 李国权1,3, 黄正文2, 庞宇3   

  1. 1. 重庆邮电大学通信与信息工程学院, 重庆 400065;
    2. 布鲁内尔大学电子与计算机工程系, 英国伦敦 UB8 3PH;
    3. 光电信息感测与传输技术重点实验室, 重庆 400065
  • 收稿日期:2020-08-20 修回日期:2021-03-26 出版日期:2021-06-25
    • 通讯作者:
    • 李国权
    • 作者简介:
    • 林金朝 男,1966年7月出生,四川蓬溪人.2001年在重庆大学获得工学博士学位.现为重庆邮电大学教授、硕士生导师,主要研究方向为医学信号处理、无线通信传输技术等;李必禄 男,1997年5月出生,四川巴中人.重庆邮电大学硕士研究生,主要研究方向为生物信号处理.
    • 基金资助:
    • 国家重点研发计划 (No.2019YFC1511300); 国家自然科学基金 (No.61671091,No.61971079); 重庆市自然科学基金面上项目 (No.cstc2019jcyj-msxmX0666); 四川省区域创新合作项目 (No.2020YFQ0025); 重庆市创新群体 (No.cstc2020jcyj-cxttX0002)

Recognition Algorithm of R Wave in ECG Based on EWT and Structure Feature Extraction

LIN Jin-zhao1,3, LI Bi-lu1,3, LI Guo-quan1,3, HUANG Zheng-wen2, PANG Yu3   

  1. 1. School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China;
    2. Department of Electronic and Computer Engineering, Brunel University London, London UB8 3PH, Britain;
    3. Chongqing Key Laboratory of Photoelectronic Information Sensing and Transmission Technology, Chongqing 400065, China
  • Received:2020-08-20 Revised:2021-03-26 Online:2021-06-25 Published:2022-06-25
    • Corresponding author:
    • LI Guo-quan
    • Supported by:
    • National Key Research and Development Program of China (No.2019YFC1511300); National Natural Science Foundation of China (No.61671091, No.61971079); General Program of Natural Science Foundation of Chongqing Municipality,  China (No.cstc2019jcyj-msxmX0666); Sichuan Province Regional Innovation Cooperation Project (No.2020YFQ0025); Chongqing Innovation Group (No.cstc2020jcyj-cxttX0002)

摘要: R波作为心电信号中最明显的特征,常作为确定心电信号其他波段的重要依据.针对现有算法识别率低的问题,提出一种基于经验小波变换和信号结构特征的R波识别算法.首先利用经验小波变换对心电信号频谱进行自适应分割,在分割区间上构造合适的小波滤波器组提取出具有紧支撑的模态分量,然后对提取出的各模态分量进行频谱分析,找出R波对应的高频分量并对其进行结构分析,从而实现R波的准确定位.仿真结果表明,所提算法对心电信号R波识别的灵敏度达到99.93%,准确率达到了99.92%,阳性准确率达到99.99%,并且算法耗时仅0.68s,对R波具有很好的识别效果.

关键词: 心电信号, R波识别, 经验小波变换, 结构特征提取

Abstract: As the most obvious feature of electrocardiogram (ECG), R wave is often used as an important basis to determine other bands of ECG. Aiming at the low recognition rate of existing algorithms, an R-wave recognition algorithm based on empirical wavelet transform and signal structure characteristics is proposed. Firstly, the empirical wavelet transform is used to adaptively segment the spectrum of ECG signal, and the appropriate wavelet filter banks are constructed in the segmentation interval to extract the tightly supported modal components. Then, the spectrum of each extracted modal component is analyzed to find out the corresponding high frequency component of R wave and analyze its structure, so as to realize the accurate positioning of R wave. The simulation results show that the sensitivity, accuracy and positive rate of the proposed algorithm for R-wave recognition of ECG signal are 99.93%, 99.92% and 99.99% respectively, aand the algorithm takes only 0.68s with a good recognition effect for R-wave.

Key words: electrocardiogram (ECG), R-wave recognition, empirical wavelet transform (EWT), structural feature extraction

中图分类号: