Automated ECG arrhythmia classification using hybrid CNN-SVM architectures

Abstract

Diagnosing cardiac diseases relies heavily on electrocardiogram (ECG) analysis, but detecting myocardial infarction-related arrhythmias remains challenging due to irregular heartbeats and signal variations. Despite advancements in machine learning, achieving both high accuracy and low computational cost for arrhythmia classification remains a critical issue. Computer-aided diagnosis systems can play a key role in early detection, reducing mortality rates associated with cardiac disorders. This study proposes a fully automated approach for ECG arrhythmia classification using deep learning and machine learning techniques to improve diagnostic accuracy while minimizing processing time. The methodology consists of three stages: 1) preprocessing, where ECG signals undergo noise reduction and feature extraction; 2) feature identification, where deep convolutional neural network (CNN) blocks, combined with data augmentation and transfer learning, extract key parameters; 3) classification, where a hybrid CNN-SVM model is employed for arrhythmia recognition. CNN-extracted features were fed into a binary support vector machine (SVM) classifier, and model performance was assessed using five-fold cross-validation. Experimental findings demonstrated that the CNN2 model achieved 85.52% accuracy, while the hybrid CNN2-SVM approach significantly improved accuracy to 97.33%, outperforming conventional methods. This model enhances classification efficiency while reducing computational complexity. The proposed approach bridges the gap between accuracy and processing speed in ECG arrhythmia classification, offering a promising solution for real-time clinical applications. Its superior performance compared to nonlinear classifiers highlights its potential for improving automated cardiac diagnosis.

Keywords

Arrhythmia / Classification / Convolutional neural networks / ECG signals / Support vector machine

Ethics approval and consent to participate

This study was conducted in accordance with the principles of the Declaration of Helsinki established by the World Medical Association and approved by the Human Ethics Committee of Bechir Hamza Children's Hospital of Tunis. All methods were conducted in compliance with relevant guidelines and regulations. Informed consent was waived by the Human Ethics Committee of Bechir Hamza Children's Hospital of Tunis.

Availability of data and materials

All data analyzed during this study are included in the published article [24]:

M.-H. Wu, E.Y. Chang, DeepQ arrhythmia database: a large-scale dataset for arrhythmia detector evaluation, in: Proc. of the 2nd Intl. Workshop on Multimedia for Personal Health and Health Care, Mountain, USA, 2017, pp. 77–80.

CRediT authorship contribution statement

Amine Ben Slama: Writing–original draft, Visualization, Formal Analysis, Investigation, Conceptualization. Yessine Amri: Writing–original draft, Visualization, Formal Analysis, Investigation, Conceptualization. Ahmed Fnaiech: Writing–original draft, Writing–review & editing. Hanene Sahli: Writing–review & editing, validation, supervision.

Declaration of competing interest

The authors declare no conflicts of interest.

References

[1] F. Coronado, S.C. Melvin, R.A. Bell, G. Zhao, Global responses to prevent, manage, and control cardiovascular diseases, Prev. Chronic Dis. 19 (2022) 1-6. 220347.

[2] E. Wilkins, L. Wilson, K. Wickramasinghe, Townsend, European Cardiovascular Disease Statistics 2017, European Heart Network, Brussels, (2017).

[3] J.C. Edelmann, D. Mair, D. Ziesel, M. Burtscher, T. Ussmueller, An ECG simulator with a novel ECG profile for physiological signals, J. Med. Eng. Technol. 42 (7) (2018) 501-509.

[4] F. Murat, O. Yildirim, M. Talo, U.B. Baloglu, Y. Demir, U.R. Acharya, Application of deep learning techniques for heartbeats detection using ECG signals-analysis and review, Comput. Biol. Med. 120 (2020) 103726.

[5] J. Venton, P.M. Harris, A. Sundar, N.A. Smith, P.J. Aston, Robustness of convolutional neural networks to physiological electrocardiogram noise, Philos. T. R. Soc. A 379 (2212) (2021) 20200262.

[6] P.M. Tripathi, A. Kumar, R. Komaragiri, M. Kumar, A review on computational methods for denoising and detecting ECG signals to detect cardiovascular diseases, Arch. Comput. Methods Eng. 29 (3) (2022) 1875-1914.

[7] A.Y. Hannun, P. Rajpurkar, M. Haghpanahi, et al., Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network, Nat. Med. 25 (1) (2019) 65-69.

[8] P. Madan, V. Singh, D.P. Singh, M. Diwakar, A. Kishor, Denoising of ECG signals using weighted stationary wavelet total variation, Biomed. Signal Process Control 73 (2022) 103478.

[9] A. Rath, D. Mishra, G. Panda, S.C. Satapathy, K.-J. Xia, Improved heart disease detection from ECG signal using deep learning based ensemble model, Sustain. Comput. Infor. 35 (2022) 100732.

[10] H.-Q. Li, Z.-F. Lin, Z.-X. An, et al., Automatic electrocardiogram detection and classification using bidirectional long short-term memory network improved by Bayesian optimization, Biomed, Signal Process. 73 (2022) 103424.

[11] S. Chabchoub, S. Mansouri, R. Ben Salah, Signal processing techniques applied to impedance cardiography ICG signals–a review, J. Med. Eng. Technol. 46 (3) (2022) 243-260.

[12] K.-Y. Cao, X.-Y. Lv, Multi-task feature fusion network for Obstructive Sleep Apnea detection using single-lead ECG signal, Measurement 202 (2022) 111787.

[13] H.-Q. Li, Z.-X. An, S.-S. Zuo, et al., Classification of electrocardiogram signals with waveform morphological analysis and support vector machines, Med. Biol. Eng. Comput. 60 (1) (2022) 109-119.

[14] A.B. Slama, H. Sahli, R. Maalmi, H. Trabelsi, ConvNet: 1D-convolutional neural networks for cardiac arrhythmia recognition using ECG signals, Trait, Signal 38 (6) (2021) 1737-1745.

[15] A.B. Slama, Ł. Lentka, A. Mouelhi, M.F. Diouani, M. Sayadi, J. Smulko, Application of statistical features and multilayer neural network to automatic diagnosis of arrhythmia by ECG signals, Metrol. Meas. Syst. 25 (1) (2018) 87-101.

[16] J.-L. Gao, H.-P. Zhang, P. Lu, Z.-M. Wang, An effective LSTM recurrent network to detect arrhythmia on imbalanced ECG dataset, J. Healthc. Eng. 2019 (2019) 6320651.

[17] L.P.A. Arts, E.L. van den Broek, The fast continuous wavelet transformation (fCWT) for real-time, high-quality, noise-resistant time–frequency analysis, Nat. Comput. Sci. 2 (1) (2022) 47-58.

[18] Q.B. Zhao, L.Q. Zhang, ECG feature extraction and classification using wavelet transform and support vector machines, in: Proc. of Intl. Conf. on Neural Networks and Brain, Beijing, China, (2005), pp. 1089-1092.

[19] S. Karimulla, D. Patra, An optimal methodology for early prediction of sudden cardiac death using advanced heart rate variability features of ECG signal, Arabian J. Sci. Eng. 49 (5) (May 2024) 6725-6741.

[20] V. Gupta, M. Mittal, Arrhythmia detection in ECG signal using fractional wavelet transform with principal component analysis, J. Inst. Eng.: Ser. Bibliogr. 101 (5) (2020) 451-461.

[21] M.A.Z. Fariha, R. Ikeura, S. Hayakawa, S. Tsutsumi, Analysis of Pan-Tompkins algorithm performance with noisy ECG signals, J. Phys. Conf. 1532 (2020) 012022.

[22] H. El-Saadawy, M. Tantawi, H.A. Shedeed, M.F. Tolba, Hybrid hierarchical method for electrocardiogram heartbeat classification, IET Signal Process. 12 (4) (2018) 506-513.

[23] P. Rajpurkar, A.Y. Hannun, M. Haghpanahi, C. Bourn, A.Y. Ng, Cardiologist-level arrhythmia detection with convolutional neural networks [Online]. Available, https://arxiv.org/abs/1707.01836, July 2017.

[24] M.H. Wu, E.Y. Chang, DeepQ arrhythmia database: a large-scale dataset for arrhythmia detector evaluation, in: Proc. of the 2nd Intl. Workshop on Multimedia for Personal Health and Health Care, Mountain, USA, (2017), pp. 77-80.

[25] M.F. Safdar, P. Pałka, R.M. Nowak, A. Al Faresi, A novel data augmentation approach for enhancement of ECG signal classification, Biomed, Signal Proces. 86 (2023) 105114. 1–11.

[26] E.H. Houssein, M. Kilany, A.E. Hassanien, ECG signals classification: a review, Int. J. Intell. Eng. Inf. 5 (4) (2017) 376-396.

[27] M. Zubair, C. Yoon, Cost-sensitive learning for anomaly detection in imbalanced ECG data using convolutional neural networks, Sensors 22 (11) (2022) 4075.

[28] B. Ganguly, A. Ghosal, A. Das, D. Das, D. Chatterjee, D. Rakshit, Automated detection and classification of arrhythmia from ECG signals using feature-induced long short-term memory network, IEEE Sensor, Letture 4 (8) (2020) 6001604.

[29] A.-Y. Chen, F. Wang, W.-H. Liu, et al., Multi-information fusion neural networks for arrhythmia automatic detection, Comput. Methods Progr. Biomed. 193 (2020) 105479.

[30] Y. Gan, J.-C. Shi, W.-M. He, F.-J. Sun, Parallel classification model of arrhythmia based on DenseNet-BiLSTM, Biocybern. Biomed. Eng. 41 (4) (2021) 1548-1560.

[31] L.-D. Fu, B.-C. Lu, B. Nie, Z.-Y. Peng, H.-Y. Liu, X.-T. Pi, Hybrid network with attention mechanism for detection and location of myocardial infarction based on 12-lead electrocardiogram signals, Sensors 20 (4) (2020) 1020.

[32] A. Sellami, H. Hwang, A robust deep convolutional neural network with batch-weighted loss for heartbeat classification, Expert Syst. Appl. 122 (2019) 75-84.

[33] S. Biswas, B.K. Panigrahi, P.K. Nayak, G. Pradhan, S. Padmanaban, A single-pole filter-assisted improved protection scheme for the TCSC-compensated transmission line connecting large-scale wind farms, IEEE J. Emerg. Select. Topics Indust. Electron. 5 (2) (2024) 346-358.

[34] S. Biswas, P.K. Nayak, B.K. Panigrahi, G. Pradhan, An intelligent fault detection and classification technique based on variational mode decomposition-CNN for transmission lines installed with UPFC and wind farm, Elec. Power Syst. Res. 223 (2023) 109526.

[35] S. Biswas, P.K. Nayak, R.S. Prakash, S. Dutta, The effect of kernels in SVM on the fault classification accuracy of a transmission line compensated with TCSC [Online]. Available, https://doi.org/10.2139/ssrn.3577488, June 2019.June 2019.