基于变权重奇异谱分析的心律不齐识别方法

doi:10.12068/j.issn.1005-3026.2022.03.001

东北大学学报（自然科学版） ›› 2022, Vol. 43 ›› Issue (3): 305-312.DOI: 10.12068/j.issn.1005-3026.2022.03.001

• 信息与控制 • 下一篇

基于变权重奇异谱分析的心律不齐识别方法

李鸿儒，任子洋，黄友鹤，于霞

(东北大学信息科学与工程学院，辽宁沈阳110819)

修回日期:2021-06-03 接受日期:2021-06-03 发布日期:2022-05-18
通讯作者: 李鸿儒
作者简介:李鸿儒(1968-)，男，内蒙古赤峰人，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(61973067， 61903071).

Recognition Method of Arrhythmia Based on Variable Weight Singular Spectrum Analysis

LI Hong-ru， REN Zi-yang， HUANG You-he， YU Xia

School of Information Science & Engineering， Northeastern University， Shenyang 110819， China.

Revised:2021-06-03 Accepted:2021-06-03 Published:2022-05-18
Contact: LI Hong-ru
About author:-
Supported by:
-

摘要/Abstract

摘要： 现有心律不齐研究多数围绕心电信号中不同频率特性成分的分离展开，而不同子序列的信息量对于最终目标决策的贡献则缺少研究与分析.为增强高贡献度子序列对于分类器的影响，提出了一种变权重奇异谱分析与深度学习结合的识别方法.通过奇异谱分析获得多个子序列，结合各个子序列的奇异值计算随机森林下的基尼系数，并将其作为权重.变权重的序列样本用于训练神经网络模型，更高效地挖掘了有用信息，进一步提高了识别精度，最终的心律不齐识别准确率为98.35%，Macro-F1为97.95%.相对于传统的定值权重，本文提出的变权重识别方法在各个性能指标上均有明显提升.

关键词: 心电图;奇异谱分析;深度学习;心律不齐识别;随机森林

Abstract: Many existing arrhythmia researches focus on the separation of different frequency characteristic components in the ECG signal. However， the contribution of different subsequences to the final target decision-making is lack of research and analysis. In order to enhance the impact of high-contribution subsequences on the classifier， a recognition method combining variable weight singular spectrum analysis and deep learning is proposed. Multiple subsequences are obtained through singular spectrum analysis.The Gini coefficient under the random forest is calculated by the singular value of each sequence and used as the weight. The sequence samples with variable weights are used to train the neural network model， which can mine useful information more efficiently and further improve the recognition accuracy. The accuracy rate of final arrhythmia recognition is 98.35%， and Macro-F1 is 97.95%. Compared with the traditional fixed weight， the proposed recognition method of variable weight has a significant improvement in various performance indicators.

Key words: electrocardiogram(ECG); singular spectrum analysis(SSA); deep learning; arrhythmia recognition; random forest

中图分类号:

TP391.5

李鸿儒，任子洋，黄友鹤，于霞. 基于变权重奇异谱分析的心律不齐识别方法[J]. 东北大学学报（自然科学版）, 2022, 43(3): 305-312.

LI Hong-ru， REN Zi-yang， HUANG You-he， YU Xia. Recognition Method of Arrhythmia Based on Variable Weight Singular Spectrum Analysis[J]. Journal of Northeastern University(Natural Science), 2022, 43(3): 305-312.

参考文献

[1]Berkaya S K，Uysal A K，Gunal E S，et al.A survey on ECG analysis［J］.Biomedical Signal Processing and Control，2018，43:216-235.
[2]Sharma M，Tan R S，Acharya U R.A novel automated diagnostic system for classification of myocardial infarction ECG signals using an optimal biorthogonal filter bank［J］.Computers in Biology and Medicine，2018，102:341-356.
[3]郑秀玉，卢瑞祥.基于FIR滤波和数学形态学的心电信号预处理算法［J］.中国医疗设备，2015，30(9):20-23.(Zheng Xiu-yu，Lu Rui-xiang.ECG signals pre-processing based on FIR filtering and mathematical morphology［J］.China Medical Device，2015，30(9):20-23.)
[4]Singh O，Sunkaria R K.ECG signal denoising via empirical wavelet transform［J］.Australasian Physical & Engineering Sciences in Medicine，2016，40(1):1-11.
[5]方明占.基于小波变换的心电信号去噪算法研究［D］.桂林:桂林电子科技大学，2012.(Fang Ming-zhan.Research on denoising algorithm of ECG signal based on wavelet transform［D］.Guilin:University of Electronic Technology，2012.)
[6]Awal M A，Mostafa S S，Ahmad M，et al.An adaptive level dependent wavelet thresholding for ECG denoising［J］.Biocybernetics & Biomedical Engineering，2014，34(4):238-249.
[7]Chaudhary M S，Kapoor R，Sharma A K.Comparison between different wavelet transforms and thresholding techniques for ECG denoising［C］//International Conference on Advances in Engineering & Technology Research.Unnao，India:IEEE，2014:1-6.
[8]Satija U，Ramkumar B，Manikandan M S.Automated ECG noise detection and classification system for unsupervised healthcare monitoring［J］.IEEE Journal of Biomedical and Health Informatics，2018，22(99):722-732.
[9]Rakshit M，Das S.An efficient ECG denoising methodology using empirical mode decomposition and adaptive switching mean filter［J］.Biomedical Signal Processing & Control，2018，40:140-148.
[10]Yucelbas S，Yucelbas C，Tezel G，et al.Automatic sleep staging based on SVD，VMD，HHT and morphological features of single-lead ECG signal［J］.Expert Systems with Applications，2018，102:193-206.
[11]Barrios M J，Romero F，Javier F，et al.A simple SSA-based denoising technique to remove ECG interference in EMG signals［J］.Biomedical Signal Processing and Control，2016，30:117-126.
[12]Mortezaee M，Mortezaie Z，Abolghasemi V.An improved SSA-based technique for EMG removal from ECG［J］.Innovation and Research in Biomedical Engineering，2019，40(1):62-68.
[13]Mukhopadhyay S K，Krishnan S.A singular spectrum analysis-based model-free electrocardiogram denoising technique［J］.Computer Methods and Programs in Biomedicine，2020，188:105304-105318.
[14]Deepthi S，Ravikumar A，Nair R V.Evaluation of classification techniques for arrhythmia screening of astronauts［J］.Procedia Technology，2016，24:1232-1239.
[15]Kandala R，Huli R D，Pawiak P，et al.Towards real-time heartbeat classification:evaluation of nonlinear morphological features and voting method［J］.Sensors，2019，19(23):5079.
[16]Yｌldｌrｌm ，Pawiak P，Tan R S，et al.Arrhythmia detection using deep convolutional neural network with long duration ECG signals［J］.Computers in Biology and Medicine，2018，102:411-420.
[17]Oh S L，Ng E Y K，Tan R S，et al.Automated beat-wise arrhythmia diagnosis using modified U-net on extended electrocardiographic recordings with heterogeneous arrhythmia types［J］.Computers in Biology and Medicine，2018，105:92-101.
[18]Xiong Z，Nash M P，Cheng E，et al.ECG signal classification for the detection of cardiac arrhythmias using a convolutional recurrent neural network［J］.Physiological Measurement，2018，39(9):094006-094015.
[19]Sellami A，Hwang H.A robust deep convolutional neural network with batch-weighted loss for heartbeat classification［J］.Expert Systems with Applications，2019，122:75-84.
[20]Yang W，Si Y，Wang D，et al.Automatic recognition of arrhythmia based on principal component analysis network and linear support vector machine［J］.Computers in Biology and Medicine，2018，101:22-32.
[21]Zhang C，Wang G，Zhao J，et al.Patient-specific ECG classification based on recurrent neural networks and clustering technique［C］//Iasted International Conference on Biomedical Engineering.Innsbruck，Austria:IEEE，2017:63-67.
[22]Faust O，Hagiwara Y，Hong T J，et al.Deep learning for healthcare applications based on physiological signals:a review［J］.Computer Methods & Programs in Biomedicine，2018，161:1-13.
[23]Yildirim ，Baloglu U B，Tan R S，et al.A new approach for arrhythmia classification using deep coded features and LSTM networks［J］.Computer Methods and Programs in Biomedicine，2019，176:121-133.
[24]Oh S L，Ng E Y K，Tan R S，et al.Automated diagnosis of arrhythmia using combination of CNN and LSTM techniques with variable length heart beats［J］.Computers in Biology & Medicine，2018，102:278-287.
[25]Tan J H，Hagiwara Y，Pang W，et al.Application of stacked convolutional and long short-term memory network for accurate identification of CAD ECG signals［J］.Computers in Biology & Medicine，2018，94:19-26.
[26]Andersen R S，Peimankar A，Puthusserypady S.A deep learning approach for real-time detection of atrial fibrillation［J］.Expert Systems with Application，2019，115:465-473.
[27]Huang J，Chen B，Yao B，et al.ECG arrhythmia classification using STFT-based spectrogram and convolutional neural network［J］.IEEE Access，2019，7(99):92871-92880.
[28]Yildirim .A novel wavelet sequences based on deep bidirectional LSTM network model for ECG signal classification［J］.Computers in Biology & Medicine，2018，96:189-202.
[29]束伟伟.基于深度学习的心电信号分类方法研究［D］.南昌:江西财经大学，2020.(Shu Wei-wei.Research on ECG signal classification method based on deep learning［D］.Nanchang:Jiangxi University of Finance and Economics，2020.)
[30]Mousavi S，Afghah F.Inter-and intra-patient ECG heartbeat classification for arrhythmia detection:asequence to sequence deep learning approach［C］//ICASSP 2019-2019 IEEE International Conference on Acoustics，Speech and Signal Processing(ICASSP).Brighton:IEEE，2019:1308-1312.［31］Shi H，Wang H，Zhang F，et al.Inter-patient heartbeat classification based on region feature extraction and ensemble classifier［J］.Biomedical Signal Processing and Control，2019，51:97-105.［32］宋新景.基于残差连接网络与度量学习的短期心电信号分类方法研究［D］.济南:山东大学，2020.(Song Xin-jing.Research on short-term ECG signal classification method based on residual connection network and metric learning［D］.Jinan:Shandong University，2020.)［33］Chazal P D，Dwyer M O，Reilly R B.Automatic classification of heartbeats using ECG morphology and heartbeat interval features［J］.IEEE Transactions on Biomedical Engineering，2004，51(7):1196-1206.［34］Kiranyaz S，Ince T，Gabbouj M.Real-time patient-specific ECG classification by 1-D convolutional neural networks［J］.IEEE Transaction on Biomedical Engineering，2016，63(3):664-675.［35］Pan J，Tompkins W J.A real-time QRS detection algorithm［J］.IEEE Transactions on Biomedical Engineering，2007，54(3):230-236.［36］Sadiq A T，Shukr N H.Classification of cardiac arrhythmia using ID3 classifier based on wavelet transform［J］.Iraqi Journal of Science，2013，54:1167-1175.［37］Sabut S，Sahoo S，Kanungo B，et al.Multiresolution wavelet transform based feature extraction and ECG classification to detect cardiac abnormalities［J］.Measurement，2017，108(108):55-66.［38］Yao Q，Wang R，Fan X，et al.Multi-class arrhythmia detection from 12-lead varied-length ECG using attention-based time-incremental convolutional neural network［J］.Information Fusion，2020，53:174-182.

基于变权重奇异谱分析的心律不齐识别方法

Recognition Method of Arrhythmia Based on Variable Weight Singular Spectrum Analysis

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	鲍喜荣，沈晓燕，张石，苏婷. 基于改进DMAS的平面波超声成像算法及其GPU实现[J]. 东北大学学报:自然科学版, 2019, 40(7): 925-931.
[2]	鲍喜荣，李玉珍，苏婷，张石. 医疗超声成像自适应波束形成算法[J]. 东北大学学报:自然科学版, 2018, 39(4): 468-473.
[3]	鲍喜荣，高浩森，历正双，张石. 基于PDChirp-Golay信号的编码激励算法[J]. 东北大学学报（自然科学版）, 2021, 42(2): 208-212.
[4]	代茵，刘维宾，董昕阳，宋雨朦. 基于注意力机制的U-Net脑脊液细胞分割[J]. 东北大学学报（自然科学版）, 2022, 43(7): 944-950.