电子测量与仪器学报

刘光达,周葛,董梦坤,胡新蕾,蔡靖,倪维广.基于FFNN和1D CNN的实时心律失常诊断系统与算法[J].电子测量与仪器学报,2021,35(3):35-42

基于FFNN和1D CNN的实时心律失常诊断系统与算法

Real time arrhythmia diagnosis system and algorithm based on FFNN and 1D CNN

DOI：

英文关键词:ECG Rpeakdetection arrhythmia MIT BIH real time system

基金项目:吉林省科技发展项目（20190303043SF）、吉林省科技发展项目（20200404205YY）资助

作者	单位
刘光达	1.吉林大学仪器科学与电气工程学院长春130026；
周葛	1.吉林大学仪器科学与电气工程学院长春130026；
董梦坤	1.吉林大学仪器科学与电气工程学院长春130026；
胡新蕾	1.吉林大学仪器科学与电气工程学院长春130026；
蔡靖	1.吉林大学仪器科学与电气工程学院长春130026；
倪维广	1.吉林大学仪器科学与电气工程学院长春130026；2.吉林大学体育学院长春130012

Author	Institution
Liu Guangda	1.College of Instrumentation and Electrical Engineering, Jilin University, Changchun 130026, China;
Zhou Ge	1.College of Instrumentation and Electrical Engineering, Jilin University, Changchun 130026, China;
Dong Mengkun	1.College of Instrumentation and Electrical Engineering, Jilin University, Changchun 130026, China;
Hu Xinlei	1.College of Instrumentation and Electrical Engineering, Jilin University, Changchun 130026, China;
Cai Jing	1.College of Instrumentation and Electrical Engineering, Jilin University, Changchun 130026, China;
Ni Weiguang	1.College of Instrumentation and Electrical Engineering, Jilin University, Changchun 130026, China;2.Physical Education College, Jilin University, Changchun 130012, China

摘要点击次数: 830

全文下载次数: 1174

中文摘要:

为了解决心律失常实时诊断的问题，设计并实现了实时心律失常诊断系统，并提出了一种基于前向反馈神经网络（FFNN）和一维卷积神经网络（1D CNN）的实时心律失常诊断算法。系统利用可穿戴的心电图（ECG)采集设备采集心电信号并实时无线传输到客户端软件进行心律失常诊断，然后将诊断结果自动上传至服务器。心律失常诊断算法以原始胸导联ECG并采用200 ms时间窗的片段作为输入，首先使用一个基于FFNN模型的分类器实时检测R波的位置，然后提取出每3个R波之间的心电序列并重采样为长度360点作为ECG_RRR特征，最后使用一个基于1D CNN模型的分类器进行实时心律失常分类。利用MIT BIH心律失常数据库中MLII导联ECG数据训练算法模型并对系统进行测试。结果表明，提出的实时心律失常诊断系统与算法具有正确率高、实时性强且易部署的特点，对于跨病人的R波位置检测查全率为980%，查准率为995%以及整体正确率为976%，对于5分类的心律失常检测正确率为915%。

英文摘要:

To solve the problem of real time arrhythmia diagnosis, this paper designs and works out a real time arrhythmia diagnosis system and proposes a real time arrhythmia diagnosis algorithm based on feed forward neural networks(FFNN) and one dimensional convolution neural network (1D CNN). The system uses a wearable ECG (electrocardiogram) acquisition device to collect ECG signals, and then wirelessly transmits the ECG to the client software in real time for arrhythmia diagnosis, and finally automatically uploads the diagnosis results to the server. The algorithm takes raw ECG signals as input and segments with a 200 ms time window, then uses a classifier based on the FFNN model to detect the position of the R peak in real time, the algorithm extracts the ECG sequence between three R peak and resamples it to a length of 360 as the ECG_RRR feature, finally uses a classifier based on the 1D CNN model for real time arrhythmia detection. This paper uses the MIT BIH database to train the algorithm model and test the system. The results show that the real time arrhythmia diagnosis system and algorithm proposed in this paper have the characteristics of high accuracy, strong real time performance and easy deployment. The recall rate, precision rate, and the overall accuracy of the system’s inter patient R peak position predictions are 980%, 995%, and 976% respectively. The overall accuracy for 5 class inter patient arrhythmia classification is 915%.

查看全文查看/发表评论下载PDF阅读器