Abstract

The diagnosis of cardiac disorders using electrocardiograms (ECGs) is critical, yet it remains susceptible to variability and often requires time-consuming manual interpretation by experts. To enhance efficiency and reduce dependency on expert analysis, the development of automated and accurate diagnostic tools is essential. This paper proposes an innovative domain adaptation network (DAN), inspired by the Wasserstein distance and generative adversarial networks (GANs). The network utilizes labeled data from multiple subjects (source domain) to improve classification performance for ECGs in a target domain. Feature extraction is conducted using a global attention module-ResNet18 (GAM-ResNet18) deep learning (DL) model with transfer learning, focusing on critical features such as the R axis, T axis, and QRS count. The proposed architecture consists of a classifier, domain discriminator, and feature extractor, and incorporates attention mechanisms along with a variance layer to enhance signal-to-feature discrimination. The model was trained and validated on both original and noise-attenuated ECG signals from the Chapman dataset. Evaluation metrics including accuracy, precision, recall, and F1-score demonstrate the superior performance of the proposed model, particularly in arrhythmia detection—a task often hindered by noise and irregular events. This model presents a promising solution for automated and efficient arrhythmia diagnosis, addressing the limitations of traditional ECG analysis. Its integration of advanced techniques such as the Wasserstein distance, GANs, attention mechanisms, and transfer learning highlights its potential to significantly improve the diagnosis of cardiac disorders in clinical settings.

Keywords

Electrocardiogram, Arrhythmia classification, Domain adaptation network, Generative adversarial network, Global attention module, Transfer learning.

References

[1] Ansari Y, Mourad O, Qaraqe K, Serpedin E. Deep learning for ECG arrhythmia detection and classification: an overview of progress for period 2017–2023. Frontiers in Physiology. 2023; 14:1-20.

[2] Daydulo YD, Thamineni BL, Dawud AA. Cardiac arrhythmia detection using deep learning approach and time frequency representation of ECG signals. BMC Medical Informatics and Decision Making. 2023; 23(1):1-14.

[3] Ebrahimi Z, Loni M, Daneshtalab M, Gharehbaghi A. A review on deep learning methods for ECG arrhythmia classification. Expert Systems with Applications: X. 2020; 7:100033.

[4] Kolhar M, Al Rajeh AM. Deep learning hybrid model ECG classification using AlexNet and parallel dual branch fusion network model. Scientific Reports. 2024; 14(1):26919.

[5] Zhang Y, Yi J, Chen A, Cheng L. Cardiac arrhythmia classification by time–frequency features inputted to the designed convolutional neural networks. Biomedical Signal Processing and Control. 2023; 79:104224.

[6] Ayar M, Isazadeh A, Gharehchopogh FS, Seyedi M. Chaotic-based divide-and-conquer feature selection method and its application in cardiac arrhythmia classification. The Journal of Supercomputing. 2022; 78:5856-82.

[7] David S, Duraipandian K, Chandrasekaran D, Pandey D, Sindhwani N, Pandey BK. Impact of blockchain in healthcare system. In Unleashing the potentials of blockchain technology for healthcare industries 2023 (pp. 37-57). Academic Press.

[8] Liu Z, Chen P. Electrocardiogram data recognition algorithm based on variable scale fusion network model. Journal of Biomedical Engineering. 2022; 39(3):570-8.

[9] Kim YK, Lee M, Song HS, Lee SW. Automatic cardiac arrhythmia classification using residual network combined with long short-term memory. IEEE Transactions on Instrumentation and Measurement. 2022; 71:1-7.

[10] Panganiban EB, Paglinawan AC, Chung WY, Paa GL. ECG diagnostic support system (EDSS): a deep learning neural network based classification system for detecting ECG abnormal rhythms from a low-powered wearable biosensors. Sensing and Bio-Sensing Research. 2021; 31:1-15.

[11] Lee H, Yoon T, Yeo C, Oh H, Ji Y, Sim S, et al. Cardiac arrhythmia classification based on one-dimensional morphological features. Applied Sciences. 2021; 11(20):1-11.

[12] Singh PN, Mahapatra RP. A novel deep learning approach for arrhythmia prediction on ECG classification using recurrent CNN with GWO. International Journal of Information Technology. 2024; 16(1):577-85.

[13] Sumalatha U, Prakasha KK, Prabhu S, Nayak VC. Deep learning applications in ecg analysis and disease detection: an investigation study of recent advances. IEEE Access. 2024: 126258-84.

[14] Xiao Q, Lee K, Mokhtar SA, Ismail I, Pauzi AL, Zhang Q, et al. Deep learning-based ECG arrhythmia classification: a systematic review. Applied Sciences. 2023; 13(8):1-25.

[15] Saber M, Abotaleb M. Arrhythmia modern classification techniques: a review. Journal of Artificial Intelligence and Metaheuristics. 2022; 1:42-53.

[16] Petmezas G, Stefanopoulos L, Kilintzis V, Tzavelis A, Rogers JA, Katsaggelos AK, et al. State-of-the-art deep learning methods on electrocardiogram data: systematic review. JMIR Medical Informatics. 2022; 10(8): e38454.

[17] Ding C, Yao T, Wu C, Ni J. Advances in deep learning for personalized ECG diagnostics: a systematic review addressing inter-patient variability and generalization constraints. Biosensors and Bioelectronics. 2024: 117073.

[18] Kim HK, Sunwoo MH. An automated cardiac arrhythmia classification network for 45 arrhythmia classes using 12-lead electrocardiogram. IEEE Access. 2024; 12:44527-38.

[19] Anand R, Lakshmi SV, Pandey D, Pandey BK. An enhanced ResNet-50 deep learning model for arrhythmia detection using electrocardiogram biomedical indicators. Evolving Systems. 2024; 15(1):83-97.

[20] Darmawahyuni A, Nurmaini S, Tutuko B, Rachmatullah MN, Firdaus F, Sapitri AI, et al. An improved electrocardiogram arrhythmia classification performance with feature optimization. BMC Medical Informatics and Decision Making. 2024; 24(1):412.

[21] Kaniraja CP, Mishra D. A deep learning framework for electrocardiogram (ECG) super resolution and arrhythmia classification. Research on Biomedical Engineering. 2024; 40(1):199-211.

[22] Soman A, Sarath R. Optimization-enabled deep convolutional neural network with multiple features for cardiac arrhythmia classification using ECG signals. Biomedical Signal Processing and Control. 2024; 92:105964.

[23] El-ghaish H, Eldele E. ECG transform: empowering adaptive ECG arrhythmia classification framework with bidirectional transformer. Biomedical Signal Processing and Control. 2024; 89:105714.

[24] Pasha MY, Sharma N, Verma VK, Choudhary A, Sharma R. Deep learning-based ECG analysis for anomaly detection and classification using DCNN. In international conference on emerging technologies and innovation for sustainability (EmergIN) 2024 (pp. 85-91). IEEE.

[25] Mahajan P, Kaul A. Optimized multi-stage sifting approach for ECG arrhythmia classification with shallow machine learning models. International Journal of Information Technology. 2024; 16(1):53-68.

[26] Gupta A, Banerjee A, Babaria D, Lotlikar K, Raut H. Prediction and classification of cardiac arrhythmia. In sentimental analysis and deep learning: proceedings of ICSADL 2022 (pp. 527-38). Springer Singapore.

[27] Hannun AY, Rajpurkar P, Haghpanahi M, Tison GH, Bourn C, Turakhia MP, et al. Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nature Medicine. 2019; 25(1):65-9.

[28] Zhang D, Yang S, Yuan X, Zhang P. Interpretable deep learning for automatic diagnosis of 12-lead electrocardiogram. Iscience. 2021; 24(4).

[29] Ullah A, Anwar SM, Bilal M, Mehmood RM. Classification of arrhythmia by using deep learning with 2-D ECG spectral image representation. Remote Sensing. 2020; 12(10):1-14.

[30] Jin Y, Li Z, Wang M, Liu J, Tian Y, Liu Y, et al. Cardiologist-level interpretable knowledge-fused deep neural network for automatic arrhythmia diagnosis. Communications Medicine. 2024; 4(1):31.

[31] Narotamo H, Dias M, Santos R, Carreiro AV, Gamboa H, Silveira M. Deep learning for ECG classification: a comparative study of 1D and 2D representations and multimodal fusion approaches. Biomedical Signal Processing and Control. 2024; 93:106141.

[32] Sraitih M, Jabrane Y, Hajjam EHA. An automated system for ECG arrhythmia detection using machine learning techniques. Journal of Clinical Medicine. 2021; 10(22):1-22.

[33] Zheng J, Zhang J, Danioko S, Yao H, Guo H, Rakovski C. A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients. Scientific Data. 2020; 7(1):1-8.

[34] Woo S, Park J, Lee JY, Kweon IS. Cbam: convolutional block attention module. In proceedings of the European conference on computer vision (ECCV) 2018 (pp. 3-19).

[35] Zhang H, Liu C, Tang F, Li M, Zhang D, Xia L, et al. Cardiac arrhythmia classification based on 3D recurrence plot analysis and deep learning. Frontiers in Physiology. 2022; 13:1-13.

[36] Murat F, Yildirim O, Talo M, Demir Y, Tan RS, Ciaccio EJ, et al. Exploring deep features and ECG attributes to detect cardiac rhythm classes. Knowledge-Based Systems. 2021; 232:107473.

[37] Hassan SU, Mohd ZMS, Abdullah TA, Husain K. Classification of cardiac arrhythmia using a convolutional neural network and bi-directional long short-term memory. Digital Health. 2022; 8:1-13.

[38] Li J, Pang SP, Xu F, Ji P, Zhou S, Shu M. Two-dimensional ECG-based cardiac arrhythmia classification using DSE-ResNet. Scientific Reports. 2022; 12(1):1-13.

[39] Zhang Y, Liu S, He Z, Zhang Y, Wang C. A CNN model for cardiac arrhythmias classification based on individual ECG signals. Cardiovascular Engineering and Technology. 2022: 548–57.

[40] Li Y, Qian R, Li K. Inter-patient arrhythmia classification with improved deep residual convolutional neural network. Computer Methods and Programs in Biomedicine. 2022; 214:1-10.

[41] Gulrajani I, Ahmed F, Arjovsky M, Dumoulin V, Courville AC. Improved training of wasserstein gans. Advances in Neural Information Processing Systems. 2017; 30.

[42] He J, Rong J, Sun L, Wang H, Zhang Y. An advanced two-step DNN-based framework for arrhythmia detection. In advances in knowledge discovery and data mining: 24th Pacific-Asia conference, Singapore, proceedings, Part II 2020 (pp. 422-34). Springer International Publishing.

[43] Katsushika S, Kodera S, Nakamoto M, Ninomiya K, Inoue S, Sawano S, et al. The effectiveness of a deep learning model to detect left ventricular systolic dysfunction from electrocardiograms. International Heart Journal. 2021; 62(6):1332-41.

[44] Nejedly P, Ivora A, Smisek R, Viscor I, Koscova Z, Jurak P, et al. Classification of ECG using ensemble of residual CNNs with attention mechanism. In computing in cardiology 2021 (pp. 1-4). IEEE.

[45] Zubair M, Yoon C. Cost-sensitive learning for anomaly detection in imbalanced ECG data using convolutional neural networks. Sensors. 2022; 22(11):1-15.

[46] Labib MI, Nahid AA. OptRPC: a novel and optimized recurrence plot-based system for ECG beat classification. Biomedical Signal Processing and Control. 2022; 72:103328.

[47] Arjovsky M, Chintala S, Bottou L. Wasserstein generative adversarial networks. In international conference on machine learning 2017 (pp. 214-23). PMLR.

[48] Madan P, Singh V, Singh DP, Diwakar M, Pant B, Kishor A. A hybrid deep learning approach for ECG-based arrhythmia classification. Bioengineering. 2022; 9(4):1-15.

Cite this article