Review of Hybrid Deep Learning Techniques For Robust ECG Signal Classification by Addressing Noise and Class Imbalance
##plugins.themes.bootstrap3.article.main##
Abstract
This review paper presents a comprehensive analysis of recent advancements in hybrid deep learning techniques for automated electrocardiogram (ECG) signal classification, with a central focus on overcoming challenges such as signal noise and class imbalance. Given the pivotal role of ECG analysis in the early detection of cardiovascular diseases, there is a growing demand for intelligent, robust, and clinically deployable systems. The paper examines hybrid model architectures that integrate Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and attention mechanisms to effectively capture both spatial and temporal features of ECG signals. Key studies are reviewed in terms of their preprocessing strategies, feature optimization methods, and noise mitigation techniques. In addition to architectural insights, the paper outlines a structured taxonomy of hybrid approaches based on model composition, preprocessing methods, imbalance handling techniques, evaluation metrics, and deployment environments. It also explores emerging application areas, including real-time monitoring systems, telecardiology platforms, and clinical decision support tools. Common research limitations—such as computational complexity, limited interpretability, and real-world generalizability—are critically discussed. The review concludes with recommendations to guide future work toward scalable, explainable, and clinically relevant deep learning frameworks, ultimately aiming to support the next generation of intelligent cardiovascular healthcare systems.
References
1. Mahel, A.S.B. et al. (2025). A multi-scale deep learning framework combining mobile ViT-ECA and LSTM for accurate ECG analysis, IEEE Access, 13, 85473–85492 doi: 10.1109/ACCESS.2025.3568026.
2. Anu, H. Rathnakara, S. & Mallikarjunaswamy, S. (2025). Enhanced ECG signal classification with CNN-LSTM networks using aquila optimization, Eng. Technol. Appl. Sci. Res., 15(3), 23461–23466. doi: 10.48084/etasr.10492.
3. Cheng, J. Zou, Q. & Zhao, Y. (2021). ECG signal classification based on deep CNN and BiLSTM. BMC Med. Inform. Decis. Mak., 21 (1) 1–12. doi: 10.1186/S12911-021-01736-Y/TABLES/5.
4. Panigrahi, A. et al. Advanced ECG signal analysis for cardiovascular disease diagnosis using AVOA optimized ensembled deep transfer learning approaches, Comput. Mater. Contin., vol. 84, no. 1, pp. 1633– 1657, 2025, doi: 10.32604/cmc.2025.063562.
5. V. Gadag et al., “Improving myocardial infarction diagnosis with Siamese network-based ECG analysis,” PLoS One, vol. 20, no. 1 January, pp. 1–28, 2025, doi: 10.1371/journal.pone.0313390.
6. A. Mahmood, M. Alhajlah, H. Dhahri, and A. Almaslukh, “Lorenz-PSO Optimized Deep Neural Network for Enhanced Phonocardiogram Classification,” IEEE Access, vol. 13, no. April, pp. 84645–84658, 2025, doi: 10.1109/ACCESS.2025.3569327.
7. D. Health and A. Nadeem, “Exploring the Significance of Synthetic ECG Signals in Atrial Fibrillation Classification,” no. May, 2025.
8. M. Najia and B. Faouzi, “An Enhanced Hybrid Model Combining CNN , BiLSTM , and Attention Mechanism for ECG Segment Classification,” 2025, doi: 10.1177/11795972251341051.
9. S. Mandala, W. Jatmiko, S. Nurmaini, A. Rizal, and Adiwijaya, “OCADN: Improving Accuracy in Multi-class Arrhythmia Detection from ECG Signals with a Hyperparameter-Optimized CNN,” IEEE Access, vol. 13, no. February, pp. 34687–34705, 2025, doi: 10.1109/ACCESS.2025.3544273.
10. P. S. Sondkar, “Machine Learning based approach for Myocardial Infarction Prediction using real time ECG Signals,” no. February, pp. 0–6, 2025, doi: 10.1109/ISML60050.2024.11007445.
11. K. Saranya, U. Karthikeyan, A. S. Kumar, A. O. Salau, and T. Tin Tin, “DenseNet-ABiLSTM: Revolutionizing Multiclass Arrhythmia Detection and Classification Using Hybrid Deep Learning Approach Leveraging PPG Signals,” Int. J. Comput. Intell. Syst., vol. 18, no. 1, 2025, doi: 10.1007/s44196-025-00765-z.
12. T. Selvapriya and V. R. Kavitha, “Hybrid Deep Learning Approach For Ecg Signal Classification Using Sd-Rom,” vol. 33, no. 8, pp. 138 145, 2024.
13. B. Aldughayfiq, F. Ashfaq, N. Z. Jhanjhi, and M. Humayun, “A Deep Learning Approach for Atrial Fibrillation Classification Using Multi- Feature Time Series Data from ECG and PPG,” Diagnostics, vol. 13, no. 14, pp. 1–25, 2023, doi: 10.3390/diagnostics13142442.
14. T. Mahmud et al., “Ensemble Deep Learning Approach for ECG-Based Cardiac Disease Detection: Signal and Image Analysis,” 2023 Int. Conf. Inf. Commun. Technol. Sustain. Dev. ICICT4SD 2023 - Proc., no. September, pp. 70–74, 2023, doi: 10.1109/ICICT4SD59951.
2023.10303625.
15. [M. S. Islam, M. N. Islam, N. Hashim, M. Rashid, B. S. Bari, and F. Al Farid, “New Hybrid Deep Learning Approach Using BiGRU-BiLSTM and Multilayered Dilated CNN to Detect Arrhythmia,” IEEE Access, vol. 10, pp. 58081–58096, 2022, doi: 10.1109/ACCESS.2022.3178710.
16. P. Madan, V. Singh, D. P. Singh, M. Diwakar, B. Pant, and A. Kishor, “A Hybrid Deep Learning Approach for ECG-Based Arrhythmia Classification,” Bioengineering, vol. 9, no. 4, pp. 1–26, 2022, doi: 10.3390/bioengineering9040152.
17. R. Jahangir, M. N. Islam, M. S. Islam, and M. M. Islam, “ECG-based heart arrhythmia classification using feature engineering and a hybrid stacked machine learning,” BMC Cardiovasc. Disord., vol. 25, no. 1, 2025, doi: 10.1186/s12872-025-04678-9.
18. A. Alsayat et al., “Enhancing cardiac diagnostics: a deep learning ensemble approach for precise ECG image classification,” J. Big Data, vol. 12, no. 1, 2025, doi: 10.1186/s40537-025-01070-4.
19. D. Kim et al., “A novel hybrid CNN-transformer model for arrhythmia detection without R-peak identification using stock well transform,” Sci. Rep., vol. 15, no. 1, pp. 1–11, 2025, doi: 10.1038/s41598-025- 92582-9.
20. S. Lamba, S. Kumar, and M. Diwakar, “FADLEC: feature extraction and arrhythmia classification using deep learning from electrocardiograph signals,” Discov. Artif. Intell., vol. 5, no. 1, 2025, doi: 10.1007/s44163-025-00290-0.
21. M. T. Ashfaq, N. Javaid, N. Alrajeh, and S. S. Ali, “An explainable AI based new deep learning solution for efficient heart disease prediction at early stages,” Evol. Syst., vol. 16, no. 1, pp. 1–26, 2025, doi: 10.1007/s12530-025-09664-2.
22. S. Vallabhuni and K. Debasis, “Hybrid deep learning for IoT-based health monitoring with physiological event extraction,” Digit. Heal., vol. 11, pp. 1–28, 2025, doi: 10.1177/20552076251337848.
23. R. Anand et al., “Winograd Transform Based Fast Detection of Heart Disease using ECG Signals and Chest X-Ray Images,” IEEE Access, vol. 13, no. April, pp. 57119–57140, 2025, doi: 10.1109/ ACCESS.2025.3555518.
24. R. Geethanjali and A. Valarmathi, “A Deep Learning based Hybrid Model for Maternal Health Risk Detection and Multifaceted Emotion Analysis in Social Networks,” Int. J. Appl. Math. Comput. Sci., vol. 34, no. 4, pp. 565–577, 2024, doi: 10.61822/amcs-2024-0038.
25. A. Huong, K. G. Tay, K. B. Gan, and X. Ngu, “Mi-Optnet: an Optimized Deep Learning Framework for Myocardial Infarction Detection,” J. Teknol., vol. 86, no. 3, pp. 115–125, 2024, doi: 10.11113/ jurnalteknologi.v86.20409.
26. G. V. Krishna et al., “Enhanced Ecg Signal Classification Using Hybrid Cnn-Transformer Models With Tuning Techniques and Genetic Algorithm Optimization,” J. Theor. Appl. Inf. Technol., vol. 102, no. 20, pp. 7510–7521, 2024.
27. S. Rotbei, G. E. Mocerino, M. S. Haleem, L. Pecchia, and A. Botta, “Frequency and Uncertainty Driven Deep Learning Approach to Segment Electrocardiogram Signals for Effective Heart Parameters Estimation,” BHI 2024 - IEEE-EMBS Int. Conf. Biomed. Heal.
Informatics, Proc., no. Dl, 2024, doi: 10.1109/BHI62660.2024. 10913758.
28. A. H. Ameen, M. A. Mohammed, and A. N. Rashid, “Enhancing Security in IoMT: A Blockchain-Based Cybersecurity Framework for Machine Learning-Driven ECG Signal Classification,” Fusion Pract. Appl., vol. 14, no. 1, pp. 221–251, 2024, doi: 10.54216/FPA.140117.
29. M. B. Terzi and O. Arikan, “Machine learning based hybrid anomaly detection technique for automatic diagnosis of cardiovascular diseases using cardiac sympathetic nerve activity and electrocardiogram,” Biomed. Tech., vol. 69, no. 1, pp. 79–109, 2024, doi:10.1515/bmt-
2022-0406.
30. A. Martinez-Velasco, L. Martinez -Villasenor, and L. Miralles-Pechuan, “Addressing Class Imbalance in Healthcare Data: Machine Learning Solutions for Age-Related Macular Degeneration and Preeclampsia,” IEEE Lat. Am. Trans. , vol. 22, no. 10, pp. 806–820, 2024, doi:10.1109/TLA.2024.10705995.
31. C. Guo, B. Yin, and J. Hu, “An Electrocardiogram Classification Using a Multiscale Convolutional Causal Attention Network,” Electron., vol. 13, no. 2, 2024, doi: 10.3390/electronics13020326.
32. G. S. Jameel, M. N. Al-Turfi, H. F. Mahdi, and S. N. Hussain, “Designing and Implementing an ECG Reader and Heart Disease Classification Device Using Deep Learning (Comprehensive Review),” Iraqi J. Comput. Sci. Math., vol. 5, no. 3, pp. 535–551, 2024, doi:
10.52866/ijcsm.2024.05.03.034.
33. I. A. C. M et al., “Development and Validation of a Class Imbalance- Resilient Cardiac Arrest Prediction Framework Incorporating Multiscale Aggregation, ICA and Explainability,” IEEE Trans. Biomed. Eng., vol. PP, no. 5, pp. 1–14, 2024, doi: 10.1109/TBME.2024.3517635.
34. A. Ogunpola, F. Saeed, S. Basurra, A. M. Albarrak, and S. N. Qasem, “Machine Learning-Based Predictive Models for Detection of
Cardiovascular Diseases,” Diagnostics, vol. 14, no. 2, 2024, doi:10.3390/diagnostics14020144.
35. W. Ji and D. Zhu, “ECG Classification Exercise Health Analysis Algorithm Based on GRU and Convolutional Neural Network,” IEEE
Access, vol. 12, no. May, pp. 59842–59850, 2024, doi: 10.1109/ACCESS.2024.3392965.
36. L. Bontinck, K. Fonteyn, T. Dhaene, and D. Deschrijver, “ECGencode: Compact and computationally efficient deep learning feature encoder for ECG signals,” Expert Syst. Appl., vol. 255, 2024, doi: 10.1016/j.eswa.2024.124775.
37. V. N. January-december et al., “Automated ECG-Based Arrhythmia Classification Using Machine Learning and Deep Convolutional Network,” vol. 3, no. 1, pp. 18–25, 2024.
38. A. S. Ba Mahel, S. Cao, K. Zhang, S. A. Chelloug, R. Alnashwan, and M. S. A. Muthanna, “Advanced integration of 2DCNN-GRU model for accurate identification of shockable life-threatening cardiac arrhythmias: a deep learning approach,” Front. Physiol., vol. 15, no. July, pp. 1– 19, 2024, doi: 10.3389/fphys.2024.1429161.
39. H. Zakaria, E. S. H. Nurdiniyah, A. M. Kurniawati, D. Naufal, and N. Sutisna, “Morphological Arrhythmia Classification Based on Inter-patient and Two Leads ECG using Machine Learning,” IEEE Access, vol. 12, no. September, pp. 147372–147386, 2024, doi: 10.1109/ ACCESS.2024.3469640.
40. K. Mallikarjunamallu and K. Syed, “Arrhythmia classification for non- experts using infinite impulse response (IIR)-filter-based machine learning and deep learning models of the electrocardiogram,” PeerJ Comput. Sci., vol. 10, 2024, doi: 10.7717/peerj-cs.1774.
41. G. Wang et al., “Convolutional Neural Network-Based ECG Signal Classification Model/ : A Study on Addressing Class Imbalance and Enhancing Model Interpretability Convolutional Neural Network-Based ECG Signal Classification Model/ : A Study on Addressing Class Imbalance and Enhancing Model Interpretability,” 2024, doi: 10.20944/ preprints202405.1290.v1.
License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.