Comparative Analysis of ECG Image Classification for Cardiac Disease Detection Using Deep Learning Approaches
Downloads
Background: Electrocardiogram (ECG) interpretation is essential for cardiac disease diagnosis. While signal-based methods have been dominant recent advances in deep learning introduced image-based classification offering new directions for automated detection.
Objectives: This study investigates the effectiveness of Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM) and a hybrid CNN-LSTM model for ECG image classification. The focus is on whether combining spatial and temporal feature learning improves accuracy compared to individual models.
Methods: ECG images were collected from public repositories. CNN and LSTM models were first applied directly to raw data but faced limitations due to noise and class imbalance. To address these issues preprocessing was introduced: Pix2Pix GAN for denoising and augmentation, and resampling with class weighting to balance classes. The models were retrained on the processed data, followed by the development of a hybrid CNN-LSTM framework combining convolutional feature extraction with sequential learning.
Results: The hybrid model achieved the best performance, with 97.89% accuracy and a macro F1-score of 0.89. Preprocessing significantly improved recognition of minority classes, highlighting the role of GAN-based denoising and balanced training.
Conclusion: A hybrid CNN-LSTM with GAN preprocessing provides a reliable and robust framework for ECG image classification. This approach offers improved generalization and could complement or enhance traditional signal-based ECG analysis in clinical practice.
Berger, L., Haberbusch, M., & Moscato, F. (2023). Generative adversarial networks in electrocardiogram synthesis: Recent developments and challenges. Artificial Intelligence in Medicine, 143, 102632.
https://doi.org/10.1016/j.artmed.2023.102632
Omarov, B., & Momynkulov, Z. (2024). Hybrid deep learning model for heart disease detection on 12-lead electrocardiograms. Procedia Computer Science, 439–444.
https://doi.org/10.1016/j.procs.2024.08.061
Qu, Z., Chen, W., & Tiwari, P. (2024). HQ-DCGAN: Hybrid quantum deep convolutional generative adversarial network approach for ECG generation. Knowledge-Based Systems, 301, 112260.
https://doi.org/10.1016/j.knosys.2024.112260
Qin, Y., et al. (2023). MVKT-ECG: Efficient single-lead ECG classification for multi-label arrhythmia by multi-view knowledge transferring. Computers in Biology and Medicine, 166, 107503.
https://doi.org/10.1016/j.compbiomed.2023.107503
R, G., et al. (2024). Enhancing heart disease prediction with reinforcement learning and data augmentation. Systems and Soft Computing, 6, 200129. https://doi.org/10.1016/j.sasc.2024.200129
Chen, C., Jin, B., Che, C., & Li, R. (2024). OSGAN: Omni-scale and global-aware ECG arrhythmia diagnostic network. Biomedical Signal Processing and Control, 96, 106602.
https://doi.org/10.1016/j.bspc.2024.106602
Tang, K., Ma, S., Sun, X., & Guo, D. (2024). Optimizing machine learning for enhanced automated ECG analysis in cardiovascular healthcare. Egyptian Informatics Journal, 28,
https://doi.org/10.1016/j.eij.2024.100578
Karthikeyani, S., Sasipriya, S., & Ramkumar, M. (2024). Cardiac arrhythmias detection framework based on higher-order spectral distribution with deep learning. Biomedical Signal Processing and Control, 92, 105997.
https://doi.org/10.1016/j.bspc.2024.105997
Safdar, M. F., Nowak, R. M., & Pałka, P. (2024). Pre-processing techniques and artificial intelligence algorithms for electrocardiogram (ECG) signals analysis: A comprehensive review. Computers in Biology and Medicine, 170, 107908. https://doi.org/10.1016/j.compbiomed.2023.107908
Cao, Q., et al. (2021). Practical fine-grained learning-based anomaly classification for ECG image. Artificial Intelligence in Medicine, 119, 102130. https://doi.org/10.1016/j.artmed.2021.102130
Alqahtani, H., Aldehim, G., Alruwais, N., Assiri, M., Alneil, A. A., & Mohamed, A. (2024). Leveraging electrocardiography signals for deep learning-driven cardiovascular disease classification model. Heliyon, 10(16), e35621.
https://doi.org/10.1016/j.heliyon.2024.e35621
Gour, A., Gupta, M., Wadhvani, R., & Shukla, S. (2024). ECG-based heart disease classification: Advancement and review of techniques. Procedia Computer Science, 1634–1648. https://doi.org/10.1016/j.procs.2024.04.155
Jiang, S., Li, D., & Zhang, Y. (2023). A deep neural network based on multi-model and multi-scale for arrhythmia classification. Biomedical Signal Processing and Control, 85, 105060. https://doi.org/10.1016/j.bspc.2023.105060
Hua, J., Zou, J., Zou, F., & Liu, J. (2024). A multi-channel ECG signal deep compressive sensing method using tree-shaped autoencoder based on multiscale feature fusion. Biomedical Signal Processing and Control, 94, 106272. https://doi.org/10.1016/j.bspc.2024.106272
Saputra, N. A., Riza, L. S., Setiawan, A., & Hamidah, I. (2024). A systematic review for classification and selection of deep learning methods. Decision Analytics Journal, 12, 100489. https://doi.org/10.1016/j.dajour.2024.100489
Premalatha, P., et al. (2024). A novel method for detecting cardiac arrhythmias in electrocardiogram signals using the ANFIS classifier. e-Prime - Advances in Electrical Engineering, Electronics and Energy, 9, 100734. https://doi.org/10.1016/j.prime.2024.100734
Alla, M. R., & Nayak, C. (2024). A robust ECG signal enhancement technique through optimally designed adaptive filters. Biomedical Signal Processing and Control, 95, 106434. https://doi.org/10.1016/j.bspc.2024.106434
Sindhu, V. S., Lakshmi, K. J., Tangellamudi, A. S., & Begum, K. G. (2023). A novel deep neural network heartbeats classifier for heart health monitoring. International Journal of Intelligent Networks, 4, 1–10.
https://doi.org/10.1016/j.ijin.2022.11.001
Li, C., et al. (2021). DeepECG: Image-based electrocardiogram interpretation with deep convolutional neural networks. Biomedical Signal Processing and Control, 69, 102824.
https://doi.org/10.1016/j.bspc.2021.102824
Omarov, B., & Momynkulov, Z. (2024). Hybrid deep learning model for heart disease detection on 12-lead electrocardiograms. Procedia Computer Science, Elsevier B.V., 439–444.
https://doi.org/10.1016/j.procs.2024.08.061
Mavaddati, S. (2025). ECG arrhythmias classification based on deep learning methods and transfer learning technique. Biomedical Signal Processing and Control, 101, 107236.
https://doi.org/10.1016/j.bspc.2024.107236
Ashtaiwi, A. A., Khalifa, T., & Alirr, O. (2024). Enhancing heart disease diagnosis through ECG image vectorization-based classification. Heliyon, 10(18), e37574.
https://doi.org/10.1016/j.heliyon.2024.e37574
Alqahtani, H., Aldehim, G., Alruwais, N., Assiri, M., Alneil, A. A., & Mohamed, A. (2024). Leveraging electrocardiography signals for deep learning-driven cardiovascular disease classification model. Heliyon, 10(16). https://doi.org/10.1016/j.heliyon.2024.e35621
Oke, O. A., & Cavus, N. (2025). A systematic review on the impact of artificial intelligence on electrocardiograms in cardiology. International Journal of Medical Informatics, 195, 105753. https://doi.org/10.1016/j.ijmedinf.2024.105753
Rahul, J., & Sharma, L. D. (2025). Advancements in AI for cardiac arrhythmia detection: A comprehensive overview. Computer Science Review, 56, 100719.
https://doi.org/10.1016/j.cosrev.2024.100719
Jafari, M., et al. (2023). Automated diagnosis of cardiovascular diseases from cardiac magnetic resonance imaging using deep learning models: A review. Computers in Biology and Medicine, 160, 106998. https://doi.org/10.1016/j.compbiomed.2023.106998
Karoui, H., Hamza, S., & ben Ayed, Y. (2024). Deep learning for cardiac diseases classification. In Lecture Notes in Computer Science (pp. 170–182). https://doi.org/10.1007/978-3-031-70816-9_14
Alqahtani, H., Aldehim, G., Alruwais, N., Assiri, M., Alneil, A. A., & Mohamed, A. (2024). Leveraging electrocardiography signals for deep learning-driven cardiovascular disease classification model. Heliyon, 10(16), e35621.
https://doi.org/10.1016/j.heliyon.2024.e35621
Ayano, Y. M., et al. (2024). Interpretable hybrid multichannel deep learning model for heart disease classification using 12-lead ECG signal. IEEE Access, 12, 94055–94080.
https://doi.org/10.1109/ACCESS.2024.3421641
Lee, J., & Shin, M. (2024). Using beat score maps with successive segmentation for ECG classification without R-peak detection. Biomedical Signal Processing and Control, 91, 105982.
https://doi.org/10.1016/j.bspc.2024.105982
Issa, M. F., et al. (2023). Heartbeat classification based on single lead-II ECG using deep learning. Heliyon, 9(7), e17974.
https://doi.org/10.1016/j.heliyon.2023.e17974
Zhang, Y., Yi, J., Chen, A., & Cheng, L. (2023). Cardiac arrhythmia classification by time–frequency features inputted to the designed convolutional neural networks. Biomedical Signal Processing and Control, 79, 104224. https://doi.org/10.1016/j.bspc.2022.104224
Hussain, M., Hameed, T., & Afzal, H. (2023). Enhancing heart disease prediction with reinforcement learning and data augmentation. Biomedical Signal Processing and Control, 80, 104340. https://doi.org/10.1016/j.bspc.2022.104340
Sharma, P., & Tripathi, S. (2022). ECGencode: Compact and computationally efficient deep learning feature encoder for ECG signals. Biomedical Signal Processing and Control, 75, 103604. https://doi.org/10.1016/j.bspc.2022.103604
G. M. Zebari, D. A. Zebari, D. Q. Zeebaree, H. Haron, A. M. Abdulazeez, and K. Yurtkan, “Efficient CNN Approach for Facial Expression Recognition,” in Journal of Physics: Conference Series, IOP Publishing Ltd, Dec. 2021.
doi: 10.1088/1742-6596/2129/1/012083.
Dias, F. M., et al. (2023). Artificial intelligence-driven screening system for rapid image-based classification of 12-lead ECG exams: A promising solution for emergency room prioritization. IEEE Access, 11, 121739–121752.
https://doi.org/10.1109/ACCESS.2023.3328538
Alsekait, D. M., et al. (2024). Heart-Net: A multi-modal deep learning approach for diagnosing cardiovascular diseases. Computers, Materials and Continua, 80(3), 3967–3990.
https://doi.org/10.32604/cmc.2024.054591
Haleem, M. S., et al. (2021). Time adaptive ECG driven cardiovascular disease detector. Biomedical Signal Processing and Control, 70, 102968. https://doi.org/10.1016/j.bspc.2021.102968
