基于深度学习的ECG信号分类

计算机与现代化 ›› 2022, Vol. 0 ›› Issue (05): 16-20.

基于深度学习的ECG信号分类

(1.青岛大学自动化学院,山东青岛266071;2.青岛大学未来研究院,山东青岛266071)

出版日期:2022-06-08 发布日期:2022-06-08
作者简介:于雁(1996—),女,山东高密人,硕士研究生,研究方向:机器学习与模式识别,E-mail: 1317189215@qq.com; 邱磊(1971—),男,江苏南京人,教授,博士,研究方向:信号处理,智慧医疗,E-mail: 45141545@qq.com。
基金资助:
国家重点研发计划项目(2020YFB1313604)

ECG Signal Classification Based on Deep Learning

(1. School of Automation, Qingdao University, Qingdao 266071, China;
2. Institute for Future, Qingdao University, Qingdao 266071, China)

Online:2022-06-08 Published:2022-06-08

摘要/Abstract

摘要： 心电图(ECG)能够实时反映心脏状态，可用于心律失常和其它心血管疾病的准确诊断。针对ECG信号采集时的噪声干扰，重构Db6小波的4级分解量并使用巴特沃斯低通滤波实现双重去噪。将降噪后的ECG信号进行R波提取，并截取P-QRS-T波片段输入到一维改进GoogLeNet模型中进行训练。一维改进GoogLeNet是原始二维GoogLeNet的优化结构，可减少网络深度并在稀疏连接中添加最大池化层和扩张卷积加大感受野，降低计算量来提高训练性能。在MIT-BIH数据集中进行实验得到99.39%的分类准确率，比一维GoogLeNet和原始GoogLeNet分别提升了0.17个百分点和0.22个百分点，训练效率均有提升。与其他先进的技术相比，心电信号分类有了显著的改进。

关键词: 心律失常分类, Db6小波, 低通滤波, 一维改进GoogLeNet, 扩张卷积

Abstract: Electrocardiogram (ECG) can reflect the state of the heart in real time, and can be used for the accurate diagnosis of arrhythmias and other cardiovascular diseases. In view of the noise interference during ECG signal acquisition, we reconstruct the fourth-order components of Db6 wavelet, then use Butterworth low pass filter to realize double denoising. Then, from denoised ECG signals to extract the R-wave, and the P-QRS-T are intercepted and input into the one-dimensional improved GoogLeNet model for training. One-dimensional improved GoogLeNet is an improved structure of the original two-dimensional GoogLeNet, which reduces the network depth and adds the maximum pooled layer and dilated convolution in the sparse connection to increase the receptive field, so as to reduce the amount of calculation and improve the training performance. Experiments on the MIT-BIH data set show that the classification accuracy is 99.39%, which is 0.17 percentage points and 0.22 percentage points higher than the one-dimensional GoogLeNet and the original GoogLeNet respectively, and the training efficiency is improved. Signal classification has a marked improvement over other advanced techniques.

Key words: classification of arrhythmias, Db6 wavelet, low pass filtering, one-dimensional improved GoogLeNet, dilated convolution

于雁, 邱磊, . 基于深度学习的ECG信号分类[J]. 计算机与现代化, 2022, 0(05): 16-20.

YU Yan, QIU Lei, . ECG Signal Classification Based on Deep Learning[J]. Computer and Modernization, 2022, 0(05): 16-20.

参考文献［28］

［1］	吴波. 基于心电信号的身份识别与心律失常识别方法研究［D］. 济南:山东大学, 2019.
［2］	WANG T, LU C H, SUN Y N, et al. Automatic ECG classification using continuous wavelet transform and convolutional neural network［J］. Entropy, 2021,23(1). DOI: 10.3390/e23010119.
［3］	马金伟,刘盛平. 心电信号识别分类算法综述［J］. 重庆理工大学学报(自然科学), 2018,32(12):122-128.
［4］	侯晓晴,仝泽友,刘晓文. 基于改进小波变换的QRS特征提取算法研究［J］. 现代电子技术, 2020,43(13):57-61.
［5］	王超超,彭勇,廖毅,等. ECG信号改进阈值函数小波去噪算法研究［J］. 电子技术与软件工程, 2020(1):74-75.
［6］	XU B X. Application of ECG signal denoising and T-wave automatic detection based on computer deep learning［J］. Journal of Physics: Conference Series, 2020,1578(1). DOI: 10.1088/1742-6596/1578/1/012007.
［7］	宋莉,孟庆建,张光玉,等. 基于波形特征和SVM的心电信号自动分类方法研究［J］. 中国医学物理学杂志, 2010,27(4):2043-2046.
［8］	吴志勇,丁香乾,许晓伟,等. 基于深度学习和模糊C均值的心电信号分类方法［J］. 自动化学报, 2018,44(10):1913-1920.
［9］	陶亮,刘宝宁,梁玮. 基于CNN-LSTM混合模型的心律失常自动检测［J］. 山东大学学报(工学版), 2021,51(3):30-36.
［10］	叶思超,徐晨华,乔清理. 基于残差网络融合模型的心律失常分类研究［J］. 陕西师范大学学报(自然科学版), 2020,48(6):10-17.
［11］	MOODY G B, MARK R G. The impact of the MIT-BIH arrhythmia database［J］. IEEE Engineering in Medicine and Biology Magazine, 2001,20(3):45-50.
［12］	BLANCO-VELASCO M, WENG B W, BARNER K E. ECG signal denoising and baseline wander correction based on the empirical mode decomposition［J］. Computers in Biology and Medicine, 2008,38(1):1-13.
［13］	卢莉蓉,王鉴,牛晓东. 基于VMD和小波阈值的ECG肌电干扰去噪处理［J］. 传感技术学报, 2020,33(6):867-873.
［14］	LIN Y D, HU Y H. Power-line interference detection and suppression in ECG signal processing［J］. IEEE Transactions on Biomedical Engineering, 2008,55(1):354-357.
［15］	AL-SALMAN W, LI Y, WEN P. Detecting sleep spindles in EEGs using wavelet fourier analysis and statistical features［J］. Biomedical Signal Processing and Control, 2019,48:80-92.
［16］	KUMNGERN M, AUPITHAK N, KHATEB F, et al. 0.5 V fifth-order Butterworth low-pass filter using multiple-input OTA for ECG applications［J］. Sensors, 2020,20(24). DOI: 10.3390/s20247343.
［17］	KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks［J］. Communications of the ACM, 2017,60(6):84-90.
［18］	SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition［J］. arXiv preprint arXiv:1409.1556, 2014.
［19］	ARORA S, BHASKARA A, GE R, et al. Provable bounds for learning some deep representations［C］// Proceedings of the 31st International Conference on Machine Learning. 2014,1:883-891.
［20］	SZEGEDY C, LIU W, JIA Y Q, et al. Going deeper with convolutions［C］// Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. 2015:1-9.
［21］	YU F, KOLTUN V. Multi-scale context aggregation by dilated convolutions［J］. arXiv preprint arXiv:1511.07122, 2015.
［22］	李楠,蔡坚勇,李科,等. 基于多Inception结构的卷积神经网络人脸识别算法［J］. 计算机系统应用, 2020,29(2):157-162.
［23］	MULVEY J M, RUSZCZYNSKI A. A new scenario decomposition method for large-scale stochastic optimization［J］. Operations Research, 1995,43(3):477-490.
［24］	刘光达,周葛,董梦坤,等. 基于FFNN和1D-CNN的实时心律失常诊断系统与算法［J］. 电子测量与仪器学报, 2021,35(3):35-42.
［25］	ZHAI X L, TIN C. Automated ECG classification using dual heartbeat coupling based on convolutional neural network［J］. IEEE Access, 2018,6:27465-27472.
［26］	杨淑莹,桂彬彬,陈胜勇. 基于小波分解和1D-GoogLeNet的心律失常检测［J］. 电子与信息学报, 2021,43(10):3018-3027.
［27］	李兴秀,唐建军,华晶. 结合CNN与双向LSTM的心律失常分类［J］. 计算机科学与探索, 2021,15(12):2353-2361.
［28］	NEELA T, NAMBURU S. ECG signal classification using capsule neural networks［J］. IET Networks, 2021,10(3):103-109.