东北大学学报:自然科学版 ›› 2019, Vol. 40 ›› Issue (11): 1539-1543.DOI: 10.12068/j.issn.1005-3026.2019.11.004

• 信息与控制 • 上一篇    下一篇

基于CNN的心冲击信号阵发性房颤自动检测方法

蒋芳芳1, 徐敬傲1, 李任1, 徐礼胜1,2   

  1. (1. 东北大学 医学与生物信息工程学院, 辽宁 沈阳110169; 2. 沈阳东软智能医疗科技研究院有限公司, 辽宁 沈阳110167)
  • 收稿日期:2019-01-22 修回日期:2019-01-22 出版日期:2019-11-15 发布日期:2019-11-05
  • 通讯作者: 蒋芳芳
  • 作者简介:蒋芳芳(1983-),女,辽宁沈阳人,东北大学讲师,博士; 徐礼胜(1975-),男,安徽安庆人,东北大学教授,博士生导师.
  • 基金资助:
    国家自然科学基金资助项目(61801104,61773110); 辽宁省科学技术基金资助项目(20170540313); 沈阳东软智能医疗科技研究院有限公司开放课题基金资助项目(NRIHTOP1801); 东北大学第十三届(2019年)大学生创新训练计划自筹项目(191188).

Automatic Detection Method of Paroxysmal Atrial Fibrillation for Ballistocardiagram Based on CNN

JIANG Fang-fang1, XU Jing-ao1, LI Ren1, XU Li-sheng1,2   

  1. 1. School of Medicine & Biological Information Engineering, Northeastern University, Shenyang 110169, China; 2. Neusoft Research of Intelligent Healthcare Technology, Co., Ltd., Shenyang 110167, China.
  • Received:2019-01-22 Revised:2019-01-22 Online:2019-11-15 Published:2019-11-05
  • Contact: XU Li-sheng
  • About author:-
  • Supported by:
    -

摘要: 阵发性房颤具有发作突然且时间短的特点,而目前其临床诊断方法——心电信号,不适于日常监护,因此,提出一种基于心冲击信号(ballistocardiogram,BCG)的非接触式房颤自动检测方法.研究不同输入数据长度与不同网络深度的匹配关系,获取应用一维卷积神经网络(convolutional neural network,CNN)检测阵发性房颤的最优组合.通过2000组数据的测试,所提模型的最佳性能为:测试准确性94.8%、敏感性97.2%、特异性92.7%,为基于BCG信号的心律失常检测与远程日常家庭监护提供了可能性.

关键词: 心冲击信号, 心电信号, 卷积神经网络, 阵发性房颤, 日常家庭监护

Abstract: Paroxysmal atrial fibrillation(PAF)is characterized by sudden onset and short duration, at present, electrocardiogram(ECG)is applied as the clinical diagnosis method, which is inconvenient for daily monitoring. Therefore, a noninvasive atrial fibrillation automatic detection method based on ballistocardiogram(BCG)was proposed. The optimal structure of one-dimensional convolutional neural network(CNN)for detecting PAF is achieved via matching different input data lengths and network depths. Through the test of 2000 sets of data, the best performances of the model proposed are: a test accuracy rate of 94.8%, a sensitivity of 97.2%, and a specificity of 92.7%, which provides the possibility of arrhythmia detection and remote daily home monitoring from BCG.

Key words: ballistocardiogram(BCG), electrocardiogram, convolutional neural network(CNN), paroxysmal atrial fibrillation(PAF), daily home monitoring

中图分类号: