基于Inception模块的CNN-BiLSTM房颤检测与心拍分类算法

doi:10.3969/j.issn.0258-8021.2024.04.007

摘要
图/表
参考文献
相关文章 (15)

全文: PDF (1043 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要心电(ECG)自动分类技术是心律不齐的一种重要辅助诊断手段。为提高动态心电异常心拍提取的准确率,提出一种基于Inception模块的CNN-BiLSTM房颤检测与心拍分类算法。首先将ECG信号分割成采样长度为1 000个采样点的心拍片段,然后利用Inception模块提取3种不同尺度的心电特征,再通过4层一维卷积神经网络(CNN)和两层双向长短期记忆神经网络(BiLSTM)来进一步提取心电特征,最后使用一层全连接网络和softmax函数实现降维和心拍分类。为了进一步提高分类准确率,采用小波降噪技术对原始ECG进行降噪。实验采用PhysioNet/Computing in Cardiology Challenge 2017数据库提供的数据,预处理后选取60 000个心拍样本进行分类,并以准确率(Acc)和F1分数(F1-score)作为评判标准来评价模型性能。实验结果表明,所建立的模型针对3类心拍(正常、房颤、其它)的分类Acc为91.38%,F1-score为91.27%,比仅使用CNN-BiLSTM组合模型(Acc为86.61%,F1-score为86.68%)分别提高了4.77%和4.59%。因此,所提出的基于Inception模块的CNN-BiLSTM房颤检测与心拍分类算法比CNN-BiLSTM的组合模型有更好的分类效果。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张耀
	刘艳君
	刘磊

关键词 ：心律失常, 心拍分类, 卷积神经网络, 长短期记忆神经网络

Abstract：Automatic ECG classification technology is an important auxiliary diagnostic method for arrhythmia. In order to improve the accuracy of abnormal dynamic ECG heartbeat extraction, an ECG beat classification algorithm based on Inception module and CNN-BiLSTM was proposed in this paper. First, the ECG signal was divided into heartbeat segments with the length of 1000 sampling points. Next, 3 different scales heartbeats were extracted by using the Inception module. The ECG features were further extracted through a 4-layer one-dimensional convolutional neural network (CNN) and a 2-layer bidirectional long short-term memory neural network (BiLSTM). At last, a 1-layer fully connected network and a softmax function were used to reduce the dimension of feature and classify the heartbeat. To improve the classification accuracy, a wavelet denoising technique was used to reduce the noise of the raw data. The data provided by the PhysioNet/Computing in Cardiology Challenge 2017 database were used in experiments. After preprocessed, 60,000 heartbeat samples were selected for classification, and the accuracy (Acc) and F1 score (F1-score) were used as the main evaluation criteria to evaluate the performance of the model. Results showed that the established model had an Acc of 91.38 % for the three types of heartbeats (normal, atrial fibrillation, and others) and F1-score was 91.27%, which was 4.77% and 4.59% higher than that of the combined model using only CNN-BiLSTM (Acc of 86.61%, F1-score of 86.68%), respectively. In conclusion, the proposed CNN-BiLSTM atrial fibrillation detection and ECG beat classification algorithm based on the Inception module has a better classification efficacy than the CNN-BiLSTM combined model.

Key words： arrhythmia ECG heartbeat classification convolutional neural network long short-term memory neural network

收稿日期: 2022-07-15

PACS:

R318

基金资助:国家自然科学基金(61673193);江苏省自然科学基金(BK20181341);中国博士后科学基金(2022M711361)

通讯作者: ^*E-mail: yjl@jiangnan.edu.cn

引用本文:

张耀, 刘艳君, 刘磊. 基于Inception模块的CNN-BiLSTM房颤检测与心拍分类算法[J]. 中国生物医学工程学报, 2024, 43(4): 447-454.
Zhang Yao, Liu Yanjun, Liu Lei. Atrial Fibrillation Detection and ECG Heartbeat Classification Algorithm Based on Inception Module and CNN-BiLSTM. Chinese Journal of Biomedical Engineering, 2024, 43(4): 447-454.

链接本文:

http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021.2024.04.007 或 http://cjbme.csbme.org/CN/Y2024/V43/I4/447

[1] Rieta JJ, Ravelli F, Sornmo L. Advances in modeling and characterization of atrial arrhythmias[J]. Biomedical Signal Processing & Control, 2013, 8(6): 956-957.
[2] Chen Yao, Wang Xiao, Abedi V. Classification of short single lead electrocardiograms (ECGs) for atrial fibrillation detection using piecewise linear spline and XGBoost[J]. Physiological Measurement, 2018, 39(10): 104006.
[3] Krijthe BP, Anton K, Benjamin EJ, et al. Projections on the number of individuals with atrial fibrillation in the European Union, from 2000 to 2060[J]. European Heart Journal, 2013, (35): 2746-2751.
[4] 刘澄玉, 杨美程, 邸佳楠, 等. 穿戴式心电:发展历程、核心技术与未来挑战[J]. 中国生物医学工程学报, 2019, 38(6): 641-652.
[5] Bruser C, Diesel J, Zink M, et al. Automatic detection of atrial fibrillation in cardiac vibration signals[J]. IEEE Journal of Biomedical and Health Informatics, 2013, 17(1): 162-171.
[6] Li Yanjun, Tang Xiaoying, Wang Ancong, et al. Probability density distribution of delta RR intervals: a novel method for the detection of atrial fibrillation[J]. Australasian Physical & Engineering Sciences in Medicine, 2017, 40(6): 1-10.
[7] Alcaraz R, Vayá C, Cervigón R, et al. Wavelet sample entropy: a new approach to predict termination of atrial fibrillation[J]. Computers in Cardiology, 2008, 33: 597-600.
[8] Alcaraz R, Rieta JJ. Application of wavelet entropy to predict atrial fibrillation progression from the surface ECG[J]. Computational and Mathematical Methods in Medicine, 2012, 2012(7): 245213.
[9] García M, Ródenas J, Alcaraz R, et al. Application of the relative wavelet energy to heart rate independent detection of atrial fibrillation[J]. Computer Methods and Programs in Biomedicine, 2016, 131(C): 157-168.
[10] Ladavich S, Ghoraani B. Rate-independent detection of atrial fibrillation by statistical modeling of atrial activity[J]. Biomedical Signal Processing & Control, 2015: 274-281.
[11] Du Xiaochuan, Rao Nini, Qian Dingyu, et al. A novel method for real-time atrial fibrillation detection in electrocardiograms using multiple parameters[J]. Annals of Noninvasive Electrocardiology, 2014, 19(3): 217-225.
[12] Ma Caiyun, Wei Shoushui, Chen Tongshuai, et al. Integration of results from convolutional neural network in a support vector machine for the detection of atrial fibrillation[J]. IEEE Transactions on Instrumentation and Measurement, 2021, 70: 1-10.
[13] 孙振. 基于机器学习的网络流量特征选择[J]. 电子测量技术, 2017, 40(7):131-136.
[14] 杨浩, 黄茂林, 蔡志鹏, 等. 融合CNN和BiLSTM的心律失常心拍分类模型[J]. 中国生物医学工程学报, 2020, 39(6): 719-726.
[15] Babaeizadeh S, Gregg RE, Nbein ED, et al. Improvements in atrial fibrillation detection for real-time monitoring[J]. Journal of Electrocardiology, 2009, 42(6): 522-526.
[16] Mohebbi M, Ghassemian H. Prediction of paroxysmal atrial fibrillation based on non-linear analysis and spectrum and bispectrum features of the heart rate variability signal[J]. Comput Methods Programs Biomed, 2012, 105(1): 40-49.
[17] Datta S, Puri C, Mukherjee A, et al. Identifying normal, AF and other abnormal ECG rhythms using a cascaded binary classifier[C]// 2017 Computing in Cardiology (CinC). Rennes: CinC, 2017: 1-4.
[18] 李全池, 黄鑫, 罗成思, 等. 融合单导联心电图传统与深度特征的常见心律失常识别方法研究[J]. 中国生物医学工程学报, 2022, 41(1): 31-40.
[19] Kiranyaz S, Ince T, Gabbouj M. Real-time patient-specific ECG classification by 1-D convolutional neural networks[J]. IEEE Transactions on Biomedical Engineering, 2016, 63(3): 664-675.
[20] Hong Shenda, Zhou Yuxi, Wu Meng, et al. Combining deep neural networks and engineered features for cardiac arrhythmia detection from ECG recordings[J]. Physiological Measurement, 2019, 40(5): 054009.
[21] Plesinger F, Nejedly P, Viscor I, et al. Parallel use of a convolutional neural network and bagged tree ensemble for the classification of Holter ECG[J]. Physiological Measurement, 2018, 018, 39(9): 094002.
[22] Cheng Shi, Tamil L S, Levine B. A mobile health system to identify the onset of paroxysmal atrial fibrillation[C]//Proceedings of the 21st International Conference on Healthcare Informatics. Dallas:IEEE, 2015: 189-192.
[23] Szegedy C, Liu Wei, Jia Yangqing, et al. Going deeper with Convolutions[C]//Proceedings of the 21st IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Boston: IEEE, 2015: 1-9.
[24] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.
[25] Ping Yongjie, Chen Chao, Wu Lu, et al. Automatic detection of atrial fibrillation based on CNN-LSTM and Shortcut connection[J]. Healthcare, 2020, 8(2): 139.
[26] Zihlmann M, Perekrestenko D, Tschannen M. Convolutional recurrent neural networks for electrocardiogram classification[C]//Proceedings of the 21st 2017 Computing in Cardiology (CinC). Rennes: CinC, 2017: 1-4.