An Empiric Analysis of Wavelet-Based Feature Extraction on Deep Learning and Machine Learning Algorithms for Arrhythmia Classification.

Ritu Singh; Navin Rajpal; Rajesh Mehta

doi:10.9781/ijimai.2020.11.005

Authors

Ritu Singh Guru Gobind Singh Indraprastha University
Navin Rajpal Guru Gobind Singh Indraprastha University
Rajesh Mehta Thapar Institute of Engineering & Technology

DOI:

https://doi.org/10.9781/ijimai.2020.11.005

Keywords:

ECG arrhythmia, Long Short Term Memory (LSTM), Support Vector Machine, Wavelet

Supporting Agencies

The authors would like to express their special gratitude to Dr. Aditya Batra, M.D., D.M (Cardiology), Holy Heart Hospital, Rohtak, Haryana, India and Dr. S.K. Gulati M.D.(Medicine), Bharat Nursing Home, Rohtak, Haryana, India and Dr. C.V. Singh, M.D. D.A. (anesthesiology), New Janta Clinic and Vidya Vision Pathology Centre, Rohtak, Haryana, India for their expert opinions and suggestions for the feature extraction of ECG data.

Abstract

The aberration in human electrocardiogram (ECG) affects cardiovascular events that may lead to arrhythmias. Many automation systems for ECG classification exist, but the ambiguity to wisely employ the in-built feature extraction or expert based manual feature extraction before classification still needs recognition. The proposed work compares and presents the enactment of using machine learning and deep learning classification on time series sequences. The two classifiers, namely the Support Vector Machine (SVM) and the Bi-directional Long Short-Term Memory (BiLSTM) network, are separately trained by direct ECG samples and extracted feature vectors using multiresolution analysis of Maximal Overlap Discrete Wavelet Transform (MODWT). Single beat segmentation with R-peaks and QRS detection is also involved with 6 morphological and 12 statistical feature extraction. The two benchmark datasets, multi-class, and binary class, are acquired from the PhysioNet database. For the binary dataset, BiLSTM with direct samples and with feature extraction gives 58.1% and 80.7% testing accuracy, respectively, whereas SVM outperforms with 99.88% accuracy. For the multi-class dataset, BiLSTM classification accuracy with the direct sample and the extracted feature is 49.6% and 95.4%, whereas SVM shows 99.44%. The efficient statistical workout depicts that the extracted feature-based selection of data can deliver distinguished outcomes compared with raw ECG data or in-built automatic feature extraction. The machine learning classifiers like SVM with knowledge-based feature extraction can equally or better perform than Bi-LSTM network for certain datasets.

Downloads

Download data is not yet available.

References

J. Jeppesen, Jesper, et al. “O-45 Automated Seizure Detection for Epilepsy Patients Using Wearable ECG-Device,” Clinical Neurophysiology, Elsevier, vol. 130, no. 7, p. e36, 2019, doi: 10.1016/j.clinph.2019.04.360.

S. Chandra, A. Sharma, G.K. Singh, “A Comparative Analysis of Performance of Several Wavelet Based ECG Data Compression Methodologies.” IRBM, 2020, doi: 10.1016/j.irbm.2020.05.004.

H. Li, D.Yuan , X. Ma , D. Cui , L. Cao L. ,“Genetic Algorithm for the Optimization of Features and Neural Networks in ECG Signals Classification.” Scientific Reports, Nature Publishing Group, vol. 7, p. 41011, 2017, doi: 10.1038/srep41011.

M.K. Islam, et al. “Study and Analysis of Ecg Signal Using Matlab &labview as Effective Tools.” International Journal of Computer and Electrical Engineering, , IACSIT Press, vol. 4, no. 3, p. 404, 2012, doi: 10.7763/IJCEE.2012.V4.522.

M. Thomas, et al. “Automatic ECG Arrhythmia Classification Using Dual Tree Complex Wavelet Based Features.” AEU-International Journal of Electronics and Communications, Elsevier, vol. 69, no. 4, pp. 715–21, 2015, doi: 10.1016/j.aeue.2014.12.013.

F. A. Elhaj, et al. “Arrhythmia Recognition and Classification Using Combined Linear and Nonlinear Features of ECG Signals.” Computer Methods and Programs in Biomedicine, Elsevier, vol. 127, pp. 52–63, 2016, doi: 10.1016/j.cmpb.2015.12.024.

R. J. Martis, et al. “ECG Beat Classification Using PCA, LDA, ICA and Discrete Wavelet Transform.” Biomedical Signal Processing and Control, Elsevier. 8, no. 5, pp. 437–48, 2013, doi: 10.1016/j.bspc.2013.01.005.

S-N. Yu, and K-T. Chou. “Integration of Independent Component Analysis and Neural Networks for ECG Beat Classification.” Expert Systems with Applications, Elsevier, vol. 34, no. 4, pp. 2841–46, 2008, doi: 10.1016/jeswa.2007.05.006.

H. M. Rai, et al. “ECG Signal Processing for Abnormalities Detection Using Multi-Resolution Wavelet Transform and Artificial Neural Network Classifier.” Measurement, Elsevier, vol. 46, no. 9, pp. 3238–46, 2013, doi: 10.1016/j.measurement.2013.05.021.

A.K. Sangaiah, M. Arumugam, G.B. Bian, “An intelligent learning approach for improving ECG signal classification and arrhythmia analysis.” Artificial Intelligence in Medicine, vol. 103, p. 101788, 2020, doi: 10.1016/j.artmed.2019.101788.

G. Wang, et al. “A Global and Updatable ECG Beat Classification System Based on Recurrent Neural Networks and Active Learning.” Information Sciences, Elsevier, vol. 501, pp. 523–42, 2019, doi: 10.1016/j.ins.2018.06.062.

B. Hou, et al. “LSTM Based Auto-Encoder Model for ECG Arrhythmias Classification.” IEEE Transactions on Instrumentation and Measurement, IEEE, 2019. doi: 10.1109/TIM.2019.2910342.

U. B. Baloglu, et al. “Classification of Myocardial Infarction with Multi-Lead ECG Signals and Deep CNN.” Pattern Recognition Letters, Elsevier, vol. 122, pp. 23–30, 2019, doi: 10.1016/j.patrec.2019.02.016.

F. Zhu, et al. “Electrocardiogram Generation with a Bidirectional LSTMCNN Generative Adversarial Network.” Scientific Reports, Nature Publishing Group, vol. 9, no. 1, pp. 1–11, 2019, doi:10.1038/s41598-019-42516-z.

Al Rahhal, et al. “Deep Learning Approach for Active Classification of Electrocardiogram Signals.” Information Sciences, Elsevier, vol. 345, pp. 340–54,2016, doi: 10.1016/j.ins.2016.01.082.

S. L. Oh, E. Y. Ng, R. S. Tan, U. R. Acharya, “Automated diagnosis of arrhythmia using combination of CNN and LSTM techniques with variable length heart beats.” Comput Biol Med, vol. 102, pp. 278–287, 2018, doi: 10.1016/j.compbiomed.2018.06.002.

A.G. Hafez, and E. Ghamry. “Geomagnetic Sudden Commencement Automatic Detection via MODWT.” IEEE Transactions on Geoscience and Remote Sensing, IEEE, vol. 51, no. 3, pp. 1547–54, 2012, doi: 10.1109/ICCES.2009.5383235.

L. Zhu, Y. Wang, Q. Fan “MODWT-ARMA model for time series prediction.” Applied Mathematical Modelling, vol. 38, no. 5-6, pp. 1859-65, 2014, doi: 10.1016/j.apm.2013.10.002.

Z. Zhang, Q. K. Telesford, C. Giusti, K.O. Lim, D. S. Bassett, “Choosing wavelet methods, filters, and lengths for functional brain network construction.” PloS one, vol. 11, no. 6, 2016, doi: 10.1371/journal.pone.0157243.

J. Chorowski, et al. “Review and Performance Comparison of SVM-and ELM-Based Classifiers.” Neurocomputing, Elsevier, vol. 128, pp. 507–16, 2014, doi: 10.1016/j.neucom.2013.08.009.

T. Chen, R. Xu, Y. He, X. Wang. “Improving Sentiment Analysis via Sentence Type Classification Using BiLSTM-CRF and CNN.” Expert Systems with Applications, Elsevier, vol. 72, pp. 221–30, 2017, doi: 10.1016/j.eswa.2016.10.065.

G.D. Clifford, et al. “AF Classification from a Short Single Lead ECG Recording: The PhysioNet/Computing in Cardiology Challenge 2017.” Computing in Cardiology (CinC), pp. 1–4, 2017, doi: 10.22489/CinC.2017.065-469.

A. L. Goldberger, et al. “PhysioBank, PhysioToolkit, and PhysioNet: Components of a New Research Resource for Complex Physiologic Signals.” Circulation, Am Heart Assoc, vol. 101, no. 23, pp. e215--e220,2000, doi: 10.1161/01.CIR.101.23.e215.

R. Singh, R. Mehta and N. Rajpal, “Efficient Wavelet Families for ECG Classification Using Neural Classifiers.” Procedia Computer Science, Elsevier, vol. 132, pp. 11–21, 2018, doi: 10.1016/j.procs.2018.05.054.

R. Singh, R. Mehta and N. Rajpal, “Wavelet and Kernel Dimensional Reduction on Arrhythmia Classification of ECG Signals.” EAI Endorsed Transactions on Scalable Information Systems: Online First, EAI, 2020, doi:10.4108/eai.13-7-2018.163095.

S. Sahoo S, B. Kanungo, S. Behera, S. Sabut, “Multiresolution wavelet transform based feature extraction and ECG classification to detect cardiac abnormalities.” Measurement, vol. 108, pp. 55–66, 2017, doi: 10.1016/j.measurement.2017.05.022.

P. Pławiak, “Novel Methodology of Cardiac Health Recognition Based on ECG Signals and Evolutionary-Neural System.” Expert Systems with Applications, Elsevier, vol. 92, pp. 334–49, 2018, doi: 10.1016/j.eswa.2017.09.022.

V. Mondéjar-Guerra, et al. “Heartbeat Classification Fusing Temporal and Morphological Information of ECGs via Ensemble of Classifiers.” Biomedical Signal Processing and Control, Elsevier, vol. 47, pp. 41–48, 2019, doi: 10.1016/j.bspc.2018.08.007.

M. Zubair, et al. “An Automated ECG Beat Classification System Using Convolutional Neural Networks.” 2016 6th International Conference on IT Convergence and Security (ICITCS), 2016, pp. 1–5, doi: 10.1109/ICITCS.2016.7740310.

U. R. Acharya, et al. “Application of Deep Convolutional Neural Network for Automated Detection of Myocardial Infarction Using ECG Signals.” Information Sciences, Elsevier, vol. 415, pp. 190–98, 2017, doi: 10.1016/j.ins.2017.06.027.

U. R. Acharya, et al. “A Deep Convolutional Neural Network Model to Classify Heartbeats.” Computers in Biology and Medicine, Elsevier, vol. 89, pp. 389–96, 2017, doi: 10.1016/j.compbiomed.2017.08.022.

A. M. Lodhi, A.N. Qureshi, U. Sharif, Z. Ashiq, “A Novel Approach Using Voting from ECG Leads to Detect Myocardial Infarction.” Proceedings of SAI Intelligent Systems Conference, 2018, pp. 337–52, doi: 10.1007/978-3-030-01057-7_27.

H. W. Lui, and K. L. Chow, “Multiclass Classification of Myocardial Infarction with Convolutional and Recurrent Neural Networks for Portable ECG Devices.” Informatics in Medicine Unlocked, Elsevier, vol. 13, pp. 26–33, 2018, doi: 10.1016/j.imu.2018.08.002.

F. Shaheen, B. Verma, M. Asafuddoula, “Impact of Automatic Feature Extraction in Deep Learning Architecture.” 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), 2016, pp. 1–8, doi: 10.1109/DICTA.2016.7797053.

YILDIRIM, “Ecg Beat Detection and Classification System Using Wavelet Transform and Online Sequential Elm.” Journal of Mechanics in Medicine and Biology, World Scientific, vol. 19, no. 01, p. 1940008, 2019, doi: 10.1142/S0219519419400086.