An Efficient Multi-Class ECG Arrhythmia Classification Framework Using Hybrid Tetrolet Based Energy and Entropy Features
DOI:
https://doi.org/10.29284/st1jy440Keywords:
Arrhythmia, ECG, Energy, Entropy, Support Vector Machine, Tetrolet Transform.Abstract
Cardiovascular diseases, the leading global cause of mortality, claim over 18 million lives annually, with arrhythmias posing acute risks especially to aging populations. Early detection via Electrocardiogram (ECG) analysis remains critical, yet challenges persist in noisy real-world signals. This study advances automated arrhythmia detection and classification by integrating statistical, chaotic, energy, entropy, and deep learning-inspired features from ECG signals. One-dimensional ECGs are transformed into two-dimensional images and decomposed using Tetrolet transforms to extract localized microstructures, yielding energy and entropy metrics from coefficients. Complementary statistical (e.g., higher-order moments) and nonlinear chaotic features (e.g., correlation dimension) are derived from raw signals, forming a compact 18-feature set optimized via feature selection. Evaluated with Support Vector Machines (SVM) on the MIT-BIH Arrhythmia Database and augmented with wearable ECG datasets, the approach tests linear, polynomial, and Radial Basis Function (RBF) kernels across Tetrolet levels. The RBF kernel excels, achieving 99.72% overall accuracy, 99.8% sensitivity, and 99.6% specificity—surpassing state-of-the-art CNN and transformer models with minimal computational overhead (under 10ms per beat on edge devices). This robust, lightweight method enables real-time deployment in wearables and clinical settings, enhancing arrhythmia screening amid rising telemedicine demands.
References
1. Jayagopi, G., & Pushpa, S. (2018). Arrhythmia classification based on combined chaotic and statistical feature extraction. Indonesian Journal of Electrical Engineering and Computer Science, 12(1), 127–136. https://doi.org/10.11591/ijeecs.v12.i1.pp127-136
2. Muntasir Azad, F., Khaled, F., & Hasan Rumman, M.R. (2018). An efficient way to convert 1D signal to 2D digital image using energy values. Master’s thesis, Department of Computer Science and Engineering, BRAC University.
3. Ceylan, M., & Canbilen, A.E. (2017). Performance comparison of Tetrolet transform and wavelet-based transforms for medical image denoising. International Journal of Intelligent Systems and Applications in Engineering, 5(4), 222–231.
4. Pengfei, L., Wang, Y., Jiangchun, H., Wang, L., Tian, Y., Zhou, T., Tianchang, L., & Jing-song, L. (2017). High-performance personalized heartbeat classification model for long-term ECG signal. IEEE Transactions on Biomedical Engineering, 64(1), 78–86.
5. Raj, S., & Ray, K.C. (2017). ECG signal analysis using DCT-based DOST and PSO optimized SVM. IEEE Transactions on Instrumentation and Measurement, 66(3), 470–478.
6. Li, H., Huan, L., Chunjiao, M., Lu, C., Xiuli, F., Chunxiao, T., & Li, E. (2016). Novel ECG signal classification based on KICA nonlinear feature extraction. Circuits, Systems, and Signal Processing, 35(4), 1187–1197.
7. Gautam, M.K., & Giri, V.K. (2015). Feature extraction and classification of ECG signal using neuro-wavelet approach. i-manager’s Journal on Digital Signal Processing, 3(4), 20–26. https://doi.org/10.26634/jdp.3.4.3708
8. Padmavathi, K., & Ramakrishna, K.S. (2015). Detection of atrial fibrillation using autoregressive modeling. International Journal of Electrical and Computer Engineering, 5(1), 64–70.
9. Iyyanarappan, A., & Tamilpavai, G. (2014). Glaucomatous image classification using wavelet-based energy features and probabilistic neural network. International Journal of Technology Enhancements and Emerging Engineering Research, 2–4.
10. Martis, R.J., Acharya, U.R.,Ray, A.K., & Chakraborty, C. (2013). Cardiac decision making using higher order spectra. Biomedical Signal Processing and Control, 8(2), 193–203.
11. Srinivasan, K., Dauwels, J., & Ramasubba Reddy, M. (2013). Multichannel EEG compression: Wavelet-based image and volumetric coding approach. IEEE Journal of Biomedical and Health Informatics, 17(1).
12. Kohli, S.S., Makwana, N., Mishra, N., & Sagar, B. (2012). Hilbert transform-based adaptive ECG R-peak detection technique. International Journal of Electrical and Computer Engineering, 2(5), 639–643.
13. Chang, T.W., Huang, Y.P., & Sandnes, F.E. (2009). Efficient entropy-based feature selection for image retrieval. Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, 2941–2946.
14. Modjtaba, R., & Reza, S. (2009). Neural networks-based diagnosis of heart arrhythmias using chaotic and nonlinear features of HRV signals. Proceedings of the International Association of Computer Science and Information Technology, 545–549.
15. Lin, C.H. (2008). Frequency-domain features for ECG beat discrimination using grey relational analysis-based classifier. Computers and Mathematics with Applications, 55, 680–690.
16. Melgani, F., & Bazi, Y. (2008). Classification of electrocardiogram signals with support vector machines and particle swarm optimization. IEEE Transactions on Information Technology in Biomedicine, 12(5), 667–677.
17. Kleiger, R.E., Stein, P.K., & Bigger, J.T. (2005). Heart rate variability: Measurement and clinical utility. Annals of Noninvasive Electrocardiology, 10, 88–101.
18. Osowski, S., Hoai, L.T., & Markiewicz, T. (2004). Support vector machine-based expert system reliable heartbeat recognition. IEEE Transactions on Biomedical Engineering, 51(4),582–589.
19. Kohler, B.U., Hennig, C., & Orglmeister, R. (2002). The principles of software QRS detection. IEEE Engineering in Medicine and Biology Magazine, 21, 42–57.
20. Moody, G.B., & Mark, R.G. (2001). The impact of the MIT-BIH arrhythmia database. IEEE Engineering in Medicine and Biology Magazine, 20(3), 45–50.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
