+91-9561854529 editorinchief.ijama@gmail.com submitpaper@ijama.in Contact Form

Deep Learning-Based ECG Signal Classification for Automated Cardiac Arrhythmia Detection

Author(s):Rajesh Kumar Anil Kumar Tiwari

Affiliation: Department of Biomedical Engineering, Maulana Azad National Institute of Technology, Bhopal, Madhya Pradesh, India Department of Electronics and Communication Engineering, National Institute of Technology, Rourkela, Odisha, India

Page No: 49-59

Volume issue & Publishing Year: Volume 3, Issue 6, June 2026

published on: 2026/06/14

Journal: International Journal of Advanced Multidisciplinary Application.(IJAMA)

ISSN NO: 3048-9350

DOI:

Download PDF

Abstract:
Cardiac arrhythmias represent a leading cause of sudden cardiac death worldwide, affecting approximately 300,000 individuals annually in India alone. Conventional manual interpretation of 12-lead electrocardiogram (ECG) signals demands considerable clinical expertise and is susceptible to inter-observer variability, motivating the development of automated classification systems capable of consistent, rapid analysis across diverse patient populations and healthcare settings. This study presents a novel hybrid deep learning architecture combining one-dimensional convolutional neural networks (1D-CNN) and bidirectional long short-term memory (BiLSTM) networks for automated multi-class arrhythmia classification from 12-lead ECG recordings. The proposed framework incorporates a multi-scale feature extraction module, an attention mechanism for lead-specific saliency weighting, and a cascaded classification head optimised for class imbalance. The model was trained and validated on a composite dataset of 27,367 annotated 12-lead ECG recordings drawn from the PhysioNet Computing in Cardiology Challenge 2020 database, covering seven clinically significant rhythm classes: normal sinus rhythm, atrial fibrillation, ventricular tachycardia, left bundle branch block, right bundle branch block, premature atrial contraction, and ST-elevation myocardial infarction. The proposed CNN-BiLSTM model achieves an overall accuracy of 98.1%, sensitivity of 97.6%, specificity of 98.5%, F1-score of 0.979, and AUC-ROC of 0.995 on the held-out test set — substantially outperforming conventional support vector machine (87.3%), random forest (89.6%), standalone 1D-CNN (93.2%), and BiLSTM (94.7%) baselines. Ablation studies confirm the independent contributions of the attention mechanism (+1.8% accuracy) and multi-lead fusion (+1.4% accuracy). The framework demonstrates robust generalisation across age groups, gender, and signal quality categories, with mean inference time of 1.2 seconds per recording on standard hardware — clinically suitable for point-of-care screening applications.

Keywords: ECG classification, cardiac arrhythmia, deep learning, convolutional neural network, bidirectional LSTM, attention mechanism, 12-lead ECG, automated diagnosis, PhysioNet, signal processing

Reference:

  • [1] Acharya, U. R., Fujita, H., Lih, O. S., Hagiwara, Y., Tan, J. H., & Adam, M. (2017). Automated detection of arrhythmias using different intervals of tachycardia ECG segments with convolutional neural network. Information Sciences, 405, 81-90.
  • [2] Chen, T. M., Huang, C. H., Shih, E. S., Hu, Y. F., & Hwang, M. J. (2020). Detection and classification of cardiac arrhythmias by a challenge-best deep learning neural network model. iScience, 23(3), 100886.
  • [3] Clifford, G. D., et al. (2017). AF classification from a short single lead ECG recording: The PhysioNet/Computing in Cardiology Challenge 2017. Computing in Cardiology, 44, 1-4.
  • [4] Hannun, A. Y., Rajpurkar, P., Haghpanahi, M., Tison, G. H., Bourn, C., Turakhia, M. P., & Ng, A. Y. (2019). Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nature Medicine, 25(1), 65-69.
  • [5] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE CVPR, 770-778.
  • [6] Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735-1780.
  • [7] Huang, J., Chen, B., Yao, B., & He, W. (2019). ECG arrhythmia classification using STFT-based spectrogram and convolutional neural network. IEEE Access, 7, 92871-92880.
  • [8] Lagerholm, M., Peterson, C., Braccini, G., Edenbrandt, L., & Sornmo, L. (2000). Clustering ECG complexes using Hermite functions and self-organizing maps. IEEE Transactions on Biomedical Engineering, 47(7), 838-848.
  • [9] Martis, R. J., Acharya, U. R., & Min, L. C. (2013). ECG beat classification using PCA, LDA, ICA and discrete wavelet transform. Biomedical Signal Processing and Control, 8(5), 437-448.
  • [10] Natarajan, A., Chang, Y., Mariani, S., Rahman, A., Boverman, G., Vij, S., & Rubin, J. (2020). A wide and deep transformer neural network for 12-lead ECG classification. Computing in Cardiology, 47, 1-4.
  • [11] Pan, J., & Tompkins, W. J. (1985). A real-time QRS detection algorithm. IEEE Transactions on Biomedical Engineering, BME-32(3), 230-236.
  • [12] Physionet. (2020). PhysioNet Computing in Cardiology Challenge 2020. Available at: https://physionetchallenges.org/2020/
  • [13] Rajpurkar, P., Hannun, A. Y., Haghpanahi, M., Bourn, C., & Ng, A. Y. (2017). Cardiologist-level arrhythmia detection with convolutional neural networks. arXiv preprint arXiv:1707.01836.
  • [14] Ribeiro, A. H., Ribeiro, M. H., Paixão, G. M., Oliveira, D. M., Gomes, P. R., Canazart, J. A., & Ribeiro, A. L. P. (2020). Automatic diagnosis of the 12-lead ECG using a deep neural network. Nature Communications, 11(1), 1760.
  • [15] Strodthoff, N., Wagner, P., Schaeffter, T., & Samek, W. (2021). Deep learning for ECG analysis: Benchmarks and insights from PTB-XL. IEEE Journal of Biomedical and Health Informatics, 25(5), 1519-1528.
  • [16] Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., ... & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30.
  • [17] World Health Organization. (2021). Cardiovascular Diseases (CVDs) Fact Sheet. WHO Press, Geneva.
  • [18] Yildirim, O., Baloglu, U. B., Tan, R. S., Ciaccio, E. J., & Acharya, U. R. (2019). A new approach for arrhythmia classification using deep coded features and LSTM networks. Computer Methods and Programs in Biomedicine, 176, 121-133.
  • [19] Zhang, Q., Zhou, D., & Zeng, X. (2017). HeartID: A multiresolution convolutional neural network for ECG-based biometric human identification in smart health applications. IEEE Access, 5, 11805-11816.
  • [20] Zhao, Z., Liaob, S., Zhang, Q., & Li, Z. (2020). Adaptive lead weighted ResNet trained with different duration signals for classifying 12-lead ECGs. Computing in Cardiology, 47, 1-4.