Detection of cardiac arrhythmias based on optimized deep features

Subject Areas :

negar janati ¹ , Mehdi Taghizadeh ² , Omid Mahdiyar ³ , Babak Gholami ⁴

1 - Islamic Azad University, Kazerun branch
2 - Faculty of Electrical and Computer, Kazerun Branch, Islamic Azad University
3 - Faculty of Engineering, Kazerun Branch, Islamic Azad University, Kazerun, Iran
4 - Department of Electrical Engineering, Kazeroun Branch, Islamic Azad University, Kazeroun, Iran

Received: 2023-11-06 Accepted : 2024-11-24 Published : 2024-11-25

Keywords: ECG signal, cardiac arrhythmias, SVM, neural network,

Abstract :

According to the World Health Organization report, cardiovascular diseases are now recognized as the leading cause of human mortality. Cardiovascular diseases, specifically vascular diseases, are the primary cause of death worldwide, accounting for 46% of mortality, according to the latest reports. On average, 200 individuals lose their lives daily due to heart disease, while 25% of cases are reversible and can be resuscitated. Rapid, timely, and accurate diagnosis, along with specialized medical care for patients with these diseases, can significantly prevent sudden death and further complications. Electrocardiogram (ECG) recording is an easy, cost-effective, and highly effective method; therefore, the use of electrocardiography and familiarity with its principles, operation, and interpretation aids in diagnosing heart diseases. The electrocardiographic signal is the most important and fundamental signal related to the heart, with minimal complexity in recording and processing, and is used to diagnose many cardiac conditions. In this article, deep learning is used to differentiate and classify cardiac arrhythmias. Using feature selection and the TFCRF weighting method, 10 deep features are extracted and input into the classifier. A neural network classifier with an accuracy of 86.99% is selected as the top classifier, effectively distinguishing arrhythmias from each other.

References:

[1] R.J. Martis, U. Rajendra Acharya and C.M. Lim, “ECG Beat Classification Using PCA, LDA, ICA and Discrete Wavelet Transform”, Biomedical Signal Processing and Control, vol. 8, pp. 437-448, 2013.
[2] U.R. Acharya, H. Fujita, V.K. Sudarshan, S.L. Oh, M. Adam, and J.E.W. Koh, “Automated detection and localization of myocardial infarction using electrocardiogram: A comparative study of different leads”, Knowledge-Based Systems, vol. 99, pp. 146-156, 2016.
[3] A. Lal, P. Kumar and S. Halder, “Heartbeat Classification Based on Deep Convolutional Neural Network”, 2023 International Conference on Networking and Communications (ICNWC), Chennai, India, 2023, pp. 1-4, doi: 10.1109/ICNWC57852.2023.10127341.
[4] E.D. Beyli, “Combining recurrent neural networks with eigenvector methods for classification of ECG beats”, Digital Signal Processing, vol. 19, issue 2, pp. 320–329, 2009.
[5] S. Chauhan and L. Vig, “Anomaly detection in ECG time signals via deep long short-term memory networks”, 2015 IEEE International Conference on Data Science and Advanced Analytics (DSAA), Paris, France, 2015, pp. 1-7, doi: 10.1109/DSAA.2015.7344872.
[6] M.K. Das, S. Ari, “ECG Beats Classification Using Mixture of Features”, International Scholarly Research Notices, vol. 2014, p. 14, 2019.
[7] C. Chen, Y. Lin, S. Lee, W. Tsai, T. Huang, Y. Liu, M. Cheng, C. Dai, “Automated ECG classification based on 1D deep learning network Methods”, Methods, vol. 202, pp. 127-135, Jun 2022.
[8] N. Ö. Özcan and F. Gürgen, "Fuzzy Support Vector Machines for ECG Arrhythmia Detection," 2010 20th International Conference on Pattern Recognition, Istanbul, Turkey, 2010, pp. 2973-2976, doi: 10.1109/ICPR.2010.728.
[9] R.D. Labati, E. Munoz, V. Piuri and R. Sassi, “Deep-ECG: Convolultional Neural Networks for ECG biometric recognition”, Pattern Recognition Letters, vol. 126, pp. 78-85, Sep. 2019.
[10] A. Sellami, H. Hwang, “A robust deep convolutional neural network with batch-weighted loss for heartbeat
classification”, Expert Systems with Applications, Vol. 122, pp. 75–84, May 2019.
[11] J. Park, K. Lee and K. Kang, “Arrhythmia detection from heartbeat using k-nearest neighbor classifier”, 2013 IEEE International Conference on Bioinformatics and Biomedicine, Shanghai, China, 2013, pp. 15-22, doi: 10.1109/BIBM.2013.6732594.
[12] P. Pławiak, “Novel Genetic Ensembles of Classifiers Applied to Myocardium Dysfunction Recognition Basedon ECG Signals”, Swarm and Evolutionary Computation, vol. 39, pp. 192-208, April 2018.
[13] Ö. Yildirim, “A novel wavelet sequences based on deep bidirectional LSTM network model for ECG signalclassification”, Computers in Biology and Medicine, vol. 96, pp. 189-202, May 2018.