计算机应用 ›› 2018, Vol. 38 ›› Issue (12): 3638-3642.DOI: 10.11772/j.issn.1001-9081.2018040843

• 应用前沿、交叉与综合 • 上一篇    下一篇

卷积神经网络在心拍类识别中的应用

原永朋1,2, 游大涛1, 渠慎明1, 武相军1, 魏梦凡1, 朱萌博1, 耿旭东1, 贾乃仁2   

  1. 1. 河南大学 软件学院, 河南 开封 475000;
    2. 深圳瑞爱心安移动心电信息服务有限公司, 广东 深圳 518101
  • 收稿日期:2018-04-23 修回日期:2018-06-14 出版日期:2018-12-10 发布日期:2018-12-15
  • 通讯作者: 游大涛
  • 作者简介:原永朋(1990-),男,河南南阳人,硕士研究生,主要研究方向:深度学习;游大涛(1981-),男,河南周口人,讲师,博士,主要研究方向:语音信号处理、机器学习;渠慎明(1982-),男,江苏丰县人,讲师,博士,主要研究方向:视频图像处理;武相军(1980-),男,河南安阳人,教授,博士,CCF会员,主要研究方向:信息安全;魏梦凡(1991-),女,河南商丘人,硕士研究生,主要研究方向:遥感应用;朱萌博(1993-),男,河南洛阳人,硕士研究生,主要研究方向:深度学习、金融工程;耿旭东(1995-),男,河南驻马店人,硕士研究生,主要研究方向:深度学习、金融工程;贾乃仁(1950-),男,上海人,副主任医师,主要研究方向:心电图记录装置、自动诊断。
  • 基金资助:
    国家自然科学基金资助项目(U1404618);河南省科技发展计划项目(172102210186,182102311066);河南省科技攻关(国际科技合作类)项目(182102410051)。

Application of convolution neural network in heart beat recognition

YUAN Yongpeng1,2, YOU Datao1, QU Shenming1, WU Xiangjun1, WEI Mengfan1, ZHU Mengbo1, GENG Xudong1, JIA Nairen2   

  1. 1. College of Software, Henan University, Kaifeng Henan 475000, China;
    2. Shenzhen Rui Ai Xin An Mobile ECG Information Service Company Limited, Shenzhen Guangdong 518101, China
  • Received:2018-04-23 Revised:2018-06-14 Online:2018-12-10 Published:2018-12-15
  • Contact: 游大涛
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (U1404618), the Science and Technology Development Program of Henan Province (172102210186, 182102311066), the Key Science and Technology (International Cooperation in Science and Technology) Project of Henan Province (182102410051).

摘要: 心电图(ECG)心拍分类对心脏疾病的临床诊断具有重要意义,但是ECG四类心拍间数据不平衡问题严重制约着心拍分类性能的提升。针对这一问题,以卷积神经网络(CNN)为基础,首先在组合四类心拍等量数据基础上构建用于表达噪声及四类心拍间共性信息的通用CNN模型,接着以通用CNN模型为基础分别在四类心拍数据上构建四个更为有效表达对应心拍类别倾向性信息的类别CNN模型,最后综合四个类别CNN模型的输出判别心拍类型。在MIT-BIH心电图数据库上的实验结果显示,该方法的平均灵敏度为99.68%、平均阳性检测率是98.58%、综合指标是99.12%,显著优于二级联合聚类法在MIT-BIH心电图数据库上的分类性能。

关键词: 心电图, 卷积神经网络, MIT-BIH数据库, 通用卷积神经网络, 类别卷积神经网络

Abstract: ElectroCardioGram (ECG) heart beat classification plays an important role in clinical diagnosis.However, there is a serious imbalance of the available data among four types of ECG, which restricts the improvement of heart beat classification performance. In order to solve this problem, a class information extracting method based on Convolutional Neural Network (CNN) was proposed. Firstly, an general CNN model based on equivalent data of four ECG types was constructed. And then based on the general CNN model, four CNN models that more effectively express the propensity information of the four heart beat categories were constructed. Finally, the outputs of the four categories of CNN models were combined to discriminate the heart beat type. The experimental results show that the average sensitivity of the proposed method is 99.68%, the average positive detection rate is 98.58%, and the comprehensive index is 99.12%; which outperform the two-stage cluster analysis method.

Key words: ElectroCardioGram (ECG), Convolutional Neural Network (CNN), MIT-BIH database, universal convolutional neural network, class-oriented convolutional neural network

中图分类号: