基于双谱和谱特征的心电信号分类方法

doi:10.16180/j.cnki.issn1007-7820.2021.05.008

摘要/Abstract

摘要：

双谱作为最低的高阶谱,具有一系列优良特性,但也存在计算量大的缺陷。文中提出了一种在双谱矩阵基础上直接提取二维特征的心电信号特征提取方法。该方法将谱平坦度、谱亮度和谱滚降度3类谱特征组成特征向量,结合基于径向基核函数的支持向量机分类方法实现ECG信号的分类识别。然后,使用MIT-BIH数据库中的ECG信号对该方法进行了验证。实验结果表明,文中提出的基于双谱的二维谱特征提取方法计算量小,准确度达93.4%,可实现对心律不齐问题的有效诊断以及对心电信号的分类。

关键词: 心电信号, 特征提取, 双谱分析, 谱平坦度, 谱亮度, 谱滚降度, 分类识别, 支持向量机

Abstract:

As the lowest high-order spectrum, bispectrum has many excellent characteristics, but it also has the defect of large calculation. Based on bispectrum matrix, a feature extraction method for two-dimensional ECG signal is proposed in the present study. The spectral flatness, spectral brightness and spectral roll off are combined to form the eigenvector, which is further combined with the support vector machine classification method founded on radial basis kernel function to realize the classification and recognition of ECG signals. The ECG signal in MIT-BIH database is applied to verify the method. Experimental result show that the two-dimensional spectrum feature extraction method based on bispectrum proposed in this study has a small amount of computation and an accuracy of 93.4%, indicating that the proposed method can effectively realize the diagnosis of arrhythmia and the classification of ECG signals.

Key words: ECG signals, feature extraction, bispectrum analysis, spectrum flatness, spectral brightness, spectral roll-off, classification recognition, support vector machine

中图分类号:

TP274

刘姝,邵杰,张颐婷,张善章. 基于双谱和谱特征的心电信号分类方法[J]. 电子科技, 2021, 34(5): 42-46.

LIU Shu,SHAO Jie,ZHANG Yiting,ZHANG Shanzhang. ECG Classification Based on Bispectrum and Spectral Features[J]. Electronic Science and Technology, 2021, 34(5): 42-46.

图/表 5

图1

图2.

表1

表2

表3

参考文献 18

[1]	Smith S, Jackson R, Pearson T, et al. Principles for national and regional guidelines on cardiovascular disease prevention[J]. Nature Clinical Practice Cardiovascular Medicine, 2006,3(9):461-470. doi: 10.1038/ncpcardio0655
[2]	Zipes D P. Clinical application of the electrocardiogram[J]. Journal of the American College of Cardiology, 2000,36(6):1746-1748. doi: 10.1016/S0735-1097(00)00971-2
[3]	官建平. 基于双谱的辐射源个体识别技术[D]. 西安:西安电子科技大学, 2014.
	Guan Jianping. Study of individual transmitter identification based on bispectrum[D]. Xi’an:Xidian University, 2014.
[4]	陶旺林, 卢选民, 刘李娟. 基于局部围线积分双谱的通信辐射源个体识别[J]. 计算机工程与应用, 2013,49(1):131-133,136.
	Tao Wanglin, Lu Xuanmin, Liu Lijuan. Identifying individual radio transmitters with selected local surrounding-line integral Bi-spectrum[J]. Computer Engineering and Applications, 2013,49(1):131-133,136.
[5]	唐贵基, 王晓龙. 基于局部均值分解和切片双谱的滚动轴承故障诊断研究[J]. 振动与冲击, 2013,32(24):83-95.
	Tang Guiji, Wang Xiaolong. Fault diagnosis of roller bearings based on local mean decomposition and slice bispectrum[J]. Journal of Vibration and Shock, 2013,32(24):83-95
[6]	马素萍, 巨志勇, 王告. 基于主成分分析与集成距离的果蔬种类识别方法[J]. 电子科技, 2019,32(11):1-6.
	Ma Suping, Ju Zhiyong, Wang Gao. Fruits and vegetables recognition based on principal component analysis and the ensemble of distances[J]. Electronic Science and Technology, 2019,32(11):1-6.
[7]	Dhir C S, Lee S Y. Discriminant independent component analysis[J]. IEEE Transactions on Neural Networks, 2011,22(6):845-857. doi: 10.1109/TNN.2011.2122266
[8]	刘明骞, 李兵兵, 吴启军. 基于矩形积分双谱和核主分量分析的电台指纹识别[J]. 西北大学学报(自然科学版), 2011,41(1):43-47.
	Liu Mingqian, Li Bingbing, Wu Qijun. The radio transmitter fingerprint identification based on SIB and KPCA[J]. Journal of Northwest University (Natural Science Edition), 2011,41(1):43-47.
[9]	Satija U, Trivedi N, Biswal G, et al. Specific emitter identification based on variational mode decomposition and spectral features in single hop and relaying scenarios[J]. IEEE Transactions on Information Forensics and Security, 2019,14(3):581-591. doi: 10.1109/TIFS.2018.2855665
[10]	Jiang J, Zhang H F, Pi D C, et al. A novel multi-module neural network system for imbalanced heartbeats classification[J]. Expert Systems with Applications:X, 2019,1(1):1-15.
[11]	Appathurai A, Carol J J, Raja C, et al. A study on ECG signal characterization and practical implementation of some ECG characterization techniques[J]. Measurement, 2019,147(1):106384-106396. doi: 10.1016/j.measurement.2019.02.040
[12]	Zhu H Y, Pan Y, Li K, et al. Method and VLSI implementation of lossy-to-lossless LTM ECG compression framework[J]. Electronics Letters, 2019,55(2):70-72. doi: 10.1049/ell2.v55.2
[13]	蓝机满. 基于径向基神经网络的粒子群表面缺陷识别算法[J]. 电子科技, 2019,32(5):92-95.
	Lan Jiman. Particle swarm optimization surface defect recognition algorithm based on radial basis neural network[J]. Electronic Science and Technology, 2019,32(5):92-95.
[14]	Yu S, Chen Y. Electrocardiogram beat classification based on wavelet transformation and probabilistic neural network[J]. Pattern Recognition Letters, 2007,28(10):1142-1150. doi: 10.1016/j.patrec.2007.01.017
[15]	Yu S, Chou K. Integration of independent component analysis and neural networks for ECG beat classification[J]. Expert Systems with Applications, 2008,34(14):2841-2846. doi: 10.1016/j.eswa.2007.05.006
[16]	Gutiérrez G, Antonio J, Morfin M, et al. DSP-based arrhythmia classification using wavelet transform and probabilistic neural network[J]. Biomedical Signal Processing & Control, 2017,32(1):44-56.
[17]	Ebrahimnezhad H, Khoshnoud S. Classification of arrhythmias using linear predictive coefficients and probabilistic neural network[J]. Applied Medical Informatics, 2013,33(3):55-62.
[18]	Lynn H, Pan S, Kim P. A deep bidirectional GRU network model for biometric electrocardiogram classification based on recurrent neural networks[J]. IEEE Access, 2019,7(1):145395-145405. doi: 10.1109/Access.6287639

实际类型	分类结果
实际类型	N	S	V	F	合计
N	30 288	1 247	1 599	863	33 997
S	187	965	99	37	1 288
V	297	119	1 643	66	2 125
F	54	18	9	311	392

实际类型	分类结果
实际类型	S	V	F	合计
S	1 207	68	13	1 288
V	111	1 963	51	2 125
F	1	7	384	392

序号	特征	总识别率/%
文献[14]	QRS复合波	89.0
文献[15]	RR间隔	92.7
文献[16]	小波	92.7
文献[17]	线性预测系数	93.1
本文	双谱	93.4
文献[18]	QRS复合波	95.1