MACHINE LEARNING AND DEEP LEARNING FOR CARDIAC ARRHYTHMIA DETECTION: A SYSTEMATIC REVIEW OF TECHNIQUES, DATASETS, AND CLINICAL TRANSLATION

Author

Pradeep Kumar Tiwari, Dr. Nidhi Mishra

Abstract

 Cardiovascular diseases (CVDs) including arrhythmias remain the leading cause of global mortality, necessitating reliable automated detection frameworks that can augment clinical decision-making. The integration of machine learning (ML) and deep learning (DL) with electrocardiogram (ECG) signal analysis has catalyzed significant advances in arrhythmia classification accuracy, clinical interpretability, and real-time monitoring capability. This paper presents a systematic review of ML and DL-based arrhythmia detection and cardiovascular disease prediction research published between 2023 and 2026. We survey the evolution from classical supervised learning algorithms — Logistic Regression, SVM, Random Forest, XGBoost — to sophisticated deep architectures including CNN-LSTM hybrids, attention-augmented BiLSTM networks, transformer-based models, and multi-scale convolutional dense networks. Benchmark datasets including MIT-BIH Arrhythmia, Kaggle CVD, and CPSC 2018 are analyzed alongside preprocessing strategies, feature engineering techniques, and class-imbalance remediation approaches. Identified research gaps encompass limited model generalizability, insufficient explainability in clinical settings, real-time deployment constraints on wearable devices, and the absence of robust benchmarking across diverse patient populations. Future research directions spanning transformer architectures, IoT-integrated monitoring, and explainable AI are discussed in detail.

Keywords

arrhythmia detection; ECG classification; machine learning; deep learning; CNN-LSTM; cardiovascular disease; MIT-BIH; attention mechanism; wavelet transform; explainable AI

Full Text:

Download Paper PDF

References

[1] He, X., Hu, C., Ma, K., et al. (2025). ECG-based arrhythmia classification using discrete wavelet transform and attention-enhanced CNN-BiGRU model. Physical and Engineering Sciences in Medicine, 48, 1995–2009.

[2] Elsheikhy, D.H., Hassan, A.S., Yhiea, N.M., et al. (2025). Lightweight deep learning architecture for multi-lead ECG arrhythmia detection. Sensors, 25(17), 5542.

[3] Islam, M.T., et al. (2024). CAT-Net: Convolution, attention, and transformer based network for single-lead ECG arrhythmia classification. Biomedical Signal Processing and Control, 93.

[4] Din, S., Qaraqe, M., et al. (2024). ECG-based cardiac arrhythmias detection through ensemble learning and fusion of deep spatial-temporal and long-range dependency features. Artificial Intelligence in Medicine, 150, 102818.

[5] Thota, V., Prajapati, H., Joshi, Y., & Rathi, S. (2025). A lightweight CNN-attention-BiLSTM architecture for multi-class arrhythmia classification on standard and wearable ECGs. CISP-BMEI 2025. arXiv:2511.08650

[6] Castel-Feced, S., Aguilar-Palacio, I., Malo, S., et al. (2025). Prediction of cardiovascular risk using machine-learning methods: sex-specific differences. Frontiers in Cardiovascular Medicine, 12, 1579947.

[7] Gnanavelu, A., Venkataramu, C., & Chintakunta, R. (2025). Cardiovascular disease prediction using machine learning metrics. Journal of Young Pharmacists, 17(1), 226–233.

[8] Tompra, K.V., Papageorgiou, G., & Tjortjis, C. (2024). Strategic machine learning optimization for cardiovascular disease prediction and high-risk patient identification. Algorithms, 17(5), 178.

[9] Naser, M.A., Majeed, A.A., et al. (2024). A review of machine learning's role in cardiovascular disease prediction: recent advances and future challenges. Algorithms, 17(2), 78.

[10] Mishra, J., & Tiwari, M. (2024). IoT-enabled ECG-based heart disease prediction using three-layer deep learning and meta-heuristic approach. Signal, Image and Video Processing, 18, 361–367.

[11] Islam, M.T., et al. (2025). An explainable deep learning framework for trustworthy arrhythmia detection from ECG signals. Scientific Reports, 15.

[12] Wu, X., & Li, Y. (2025). A novel method of Swin Transformer with time-frequency characteristics for ECG-based arrhythmia detection. Frontiers in Physiology. PMC11193364.

[13] Singh, P.N., & Mahapatra, R.P. (2024). A novel deep learning approach for arrhythmia prediction on ECG classification using recurrent CNN with GWO. International Journal of Information Technology, 16(1), 577–585.

[14] Frontiers in AI. (2025). Machine learning techniques for improved prediction of cardiovascular diseases using integrated healthcare data. Frontiers in Artificial Intelligence, 8.

[15] Frontiers in AI. (2026). Web-based cardiovascular disease risk prediction using machine learning. Frontiers in Artificial Intelligence.

[16] Frontiers in Cardiovascular Medicine. (2026). Evaluation of coronary heart disease risk prediction using stacked ensemble and SHAP. Frontiers in Cardiovascular Medicine.

[17] Chen, Z., Yang, D., Cui, T., et al. (2024). A novel imbalanced dataset mitigation method and ECG classification model based on combined 1D_CBAM-autoencoder and lightweight CNN. Biomedical Signal Processing and Control, 87, 105437.

[18] Barfungpa, S.P., et al. (2024). SMOTE-based hybrid deep learning network for predicting patient survival in CVD. Frontiers in Artificial Intelligence, 7.

[19] Diagnostics. (2025). Enhanced prediction of cardiovascular disease through integrated machine learning models. Diagnostics, 16(10), 1572.

[20] Mavaddati, S. (2025). ECG arrhythmias classification based on deep learning methods and transfer learning technique. Biomedical Signal Processing and Control, 101, 107236.

[21] Ait Bourkha, M.E., et al. (2025). Optimized wavelet scattering network and CNN for ECG heartbeat classification from MIT-BIH arrhythmia database. International Journal of Advanced Computer Science and Applications, 16.

[22] Kailan, S.L., et al. (2025). Efficient ECG classification based on machine learning and feature selection algorithm for IoT-5G enabled health monitoring systems. International Journal of Intelligent Engineering Systems, 18.

[23] JMIR Cardio. (2025). Machine learning model for predicting coronary heart disease risk: development and validation. JMIR Cardio, 9, e68066.

[24] Mishra, J., & Tiwari, M. (2024). Heart disease prediction using ECG-based lightweight system in IoT based on meta-heuristic approach. PMC11636128.

[25] Wang, H., Yin, Z., Sun, H., & Guo, X. (2024). Enhancing vital sign monitoring with reinforcement learning and wavelet analysis in sleep disorders. IEEE ICME 2024.