心电信号的卷积神经网络二分类方法  

Convolutional neural network binary classification method for electrocardiogram signal

作　　者：沈国皓 鲁昌华[1] 王涛[1] 孙怡宁[2] 蒋文刚 Shen Guohao;Lu Changhua;Wang Tao;Sun Yining;Jiang Wengang(School of Computer and Information of Hefei University of Technology,Hefei 230000,China;Institute of Intelligent Machines,Chinese Academy of Sciences,Hefei 230031,China;Huangshan Scenic Spot Management Committee,Huangshan 245800,China)

机构地区：[1]合肥工业大学计算机与信息学院,合肥230000 [2]中国科学院合肥智能机械研究所,合肥230031 [3]黄山风景区管理委员会,黄山245800

出　　处：《电子测量与仪器学报》2021年第3期43-48,共6页Journal of Electronic Measurement and Instrumentation

基　　金：中科院STS重大项目(Grant:KFJ-STS-ZDTP-079)资助。

摘　　要：心电信号直观地反映心脏生理电活动,在诊断心血管疾病方面有重要的参考价值。提出了一种卷积神经网络的心电信号二分类方法,网络卷积层使用不同卷积核,最大限度的利用局部特征进行分类,对异常心拍进行检测。使用麻省理工学院提出的MIT-BIH心律失常数据库对该方法进行验证。通过混淆矩阵计算性能指标,运用交叉验证与3种传统机器学习方法对比。实验表明,相较于准确性能最高的支持向量机二分类方法,模型准确率可达96.86%,提升了3.39%。该方法简化了特征提取过程,充分提高了异常心拍检测的准确性。The electrocardiogram signal(ECG) intuitively reflects the physiologically electrical activities of heart, and has important reference value in diagnosing heart diseases. In this paper, we proposed a kind of two-class classification method for ECG signals using convolutional neural networks. The network convolution layer used different convolution kernels to maximize the use of local features for classification and detection of abnormal heart beats. The method has utilized the MIT-BIH Arrhythmia Database proposed by Massachusetts Institute of Technology. Calculating performance metrics through confusion matrix and applying cross-validation against three traditional machine learning methods, experiments show that the model accuracy rate can even reach 96.86%, which increases 3.39%, compared with the support vector machine dichotomy method with the highest accuracy performance. This method simplified the feature extraction process and fully improved the accuracy of abnormal heartbeat detection.

关 键 词：心电信号 卷积神经网络 二分类 异常心拍检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] TN91[自动化与计算机技术—计算机科学与技术]