基于CNN和LSTM融合心肌梗死检测和识别  

DETECTION AND LOCATION OF MYOCARDIAL INFARCTIONBASED ON FUSION OF CNN AND LSTM NETWORK

作　　者：张宏坡[1,2] 王震 董忠仁[1,2] 逯鹏 Zhang Hongpo;Wang Zhen;Dong Zhongren;Lu Peng(School of Information Engineering,Zhengzhou University,Zhengzhou 450001,Henan,China;Cooperative Innovation Center of Internet Healthcare,Zhengzhou University,Zhengzhou 450052,Henan,China;School of Electrical Engineering,Zhengzhou University,Zhengzhou 450001,Henan,China)

机构地区：[1]郑州大学信息工程学院,河南郑州450001 [2]郑州大学互联网医疗与健康服务河南省协同创新中心,河南郑州450052 [3]郑州大学电气工程学院,河南郑州450001

出　　处：《计算机应用与软件》2023年第11期201-207,272,共8页Computer Applications and Software

基　　金：河南省科技惠民计划项目(182207310002);兵团科协重点科技攻关项目(2018AB017);教育部科技发展中心“云数融合科教创新”基金资助项目(2017A11017)。

摘　　要：心肌梗死是由于血栓引起冠状动脉阻塞导致组织血流中断而引起的心肌损伤,是威胁人类生命健康的心血管疾病之一,便携式心电图设备发展使得及时检测和监测心肌梗死的心电信号成为可能。针对此问题,提出一种新的基于卷积神经网络和长短期记忆网络的算法,利用导联Ⅱ心电图来检测和定位心肌梗死。使用CAM技术对卷积层提取的心电信号特征进行可视化分析。模型在PTB诊断数据库进行验证,并通过十折交叉验证防止过拟合,心肌梗死检测的准确率和F1值分别为99.99%和99.99%,在心肌梗死的定位上的准确率和F1值分别为99.84%和99.84%。Myocardial infarction is caused by coronary artery thrombosis obstruction that leads to interrupt blood flow and cause myocardial injury.It is one of the cardiovascular diseases that threaten human life and health.The development of portable electrocardiogram equipment makes it possible to detect and monitor the ECG signals of myocardial infarction in time.Aimed at this problem,a new algorithm based on convolutional neural networks and long short-term memory networks is proposed.It used leadⅡECG to detect and locate myocardial infarction.The CAM technology was used to visualize the ECG signal features extracted by the convolutional layer.The proposed model was verified on the PTB diagnostic database and through 10-fold cross validation to prevent overfitting.The accuracy and F1 value of myocardial infarction detection were 99.99%and 99.99%,respectively.The accuracy and F1 value of myocardial infarction localization were 99.84%and 99.84%,respectively.

关 键 词：心肌梗死 卷积神经网络 长短时记忆网络 心电图 深度学习 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术] TP183