Convolutional neural network for classification of eight types of arrhythmia using 2D time-frequency feature map from standard 12-lead electrocardiogram

Da Un Jeong, Ki Moo Lim, Da Un Jeong, Ki Moo Lim

Abstract

Electrocardiograms (ECGs) are widely used for diagnosing cardiac arrhythmia based on the deformation of signal shapes due to changes in various heart diseases. However, these abnormal signs may not be observed in some 12 ECG channels, depending on the location, the heart shape, and the type of cardiac arrhythmia. Therefore, it is necessary to closely and comprehensively observe ECG records acquired from 12 channel electrodes to diagnose cardiac arrhythmias accurately. In this study, we proposed a clustering algorithm that can classify persistent cardiac arrhythmia as well as episodic cardiac arrhythmias using the standard 12-lead ECG records and the 2D CNN model using the time-frequency feature maps to classify the eight types of arrhythmias and normal sinus rhythm. The standard 12-lead ECG records were provided by China Physiological Signal Challenge 2018 and consisted of 6877 patients. The proposed algorithm showed high performance in classifying persistent cardiac arrhythmias; however, its accuracy was somewhat low in classifying episodic arrhythmias. If our proposed model is trained and verified using more clinical data, we believe it can be used as an auxiliary device for diagnosing cardiac arrhythmias.

Conflict of interest statement

The authors declare no competing interests.

© 2021. The Author(s).

Figures

Figure 1
Figure 1
Confusion matrix for the proposed model.
Figure 2
Figure 2
Performance curves for the proposed model. (a) Receiver operating characteristics (ROC) curves; (b) Precision-recall curves.
Figure 3
Figure 3
The workflow of the proposed model. (a) The workflow for the entire process; (b) the workflow for the clustering algorithm.
Figure 4
Figure 4
Time–frequency feature map. (a) Mimetic diagram of the time–frequency feature map; (b) a time–frequency feature map generated from the proposed algorithm.
Figure 5
Figure 5
The structure of the 2D CNN model.

References

    1. Rajendra Acharya U, Suri JS, Spaan JAE, Krishnan SM. Advances in cardiac signal processing. Adv. Cardiac Signal Process. 2007 doi: 10.1007/978-3-540-36675-1.
    1. Ashley EA, Niebauer J. Cardiology Explained. Andrew Ward; 2004.
    1. Hagiwara Y, et al. Computer-aided diagnosis of atrial fibrillation based on ECG Signals: A review. Inf. Sci. 2018;467:99–114. doi: 10.1016/j.ins.2018.07.063.
    1. Uhm J-S, et al. First-degree atrioventricular block is associated with advanced atrioventricular block, atrial fibrillation and left ventricular dysfunction in patients with hypertension. J. Hypertens. 2014;32:1115–1120. doi: 10.1097/HJH.0000000000000141.
    1. Dale D. Cardiology—Rapid Interpretation of EKG's. Cover Publishing Company; 2000.
    1. Fred, K. ECG Interpretation: From Pathophysiology to Clinical Application, Vol. 5, No. 2 (Springer Nature, 2020).
    1. Krasteva, V., Jekova, I. & Christov, I. Automatic detection of premature atrial contractions in the electrocardiogram. 9–10 (2006).
    1. Inan OT, Giovangrandi L, Kovacs GTA. Robust neural-network-based classification of premature ventricular contractions using wavelet transform and timing interval features. IEEE Trans. Biomed. Eng. 2006;53:2507–2515. doi: 10.1109/TBME.2006.880879.
    1. Ullah A, Anwar SM, Bilal M, Mehmood RM. Classification of arrhythmia by using deep learning with 2-D ECG spectral image representation. Remote Sens. 2020;12:1–14.
    1. Shi L, Yang C, Zhang J, Li H. Recognition of ST segment of electrocardiogram based on wavelet transform. Life Sci. J. 2007;4:90–93.
    1. Datta S, et al. Identifying normal, AF and other abnormal ECG rhythms using a cascaded binary classifier. Comput. Cardiol. 2017;2010(44):1–4.
    1. Baek YS, Lee SC, Choi W, Kim DH. A new deep learning algorithm of 12-lead electrocardiogram for identifying atrial fibrillation during sinus rhythm. Sci. Rep. 2021;11:1–10. doi: 10.1038/s41598-020-79139-8.
    1. Jadhav, S. M., Nalbalwar, S. L. & Ghatol, A. Artificial Neural Network based cardiac arrhythmia classification using ECG signal data. In ICEIE 2010—2010 Int. Conf. Electron. Inf. Eng. Proc. Vol. 1, 228–231 (2010).
    1. Ribeiro AH, et al. Automatic diagnosis of the 12-lead ECG using a deep neural network. Nat. Commun. 2020;11:1–9.
    1. Mostayed, A., Luo, J., Shu, X. & Wee, W. Classification of 12-Lead ECG signals with bi-directional LSTM network. arXiv 1–16 (2018).
    1. Chen TM, Huang CH, Shih ESC, Hu YF, Hwang MJ. Detection and classification of cardiac arrhythmias by a challenge-best deep learning neural network model. iScience. 2020;23:1–9.
    1. Lin, Y. C., Lee, Y. C., Tsai, W. C., Beh, W. K. & Wu, A. Y. A. Explainable deep neural network for identifying cardiac abnormalities using class activation map. In Computing in Cardiology Vol. 2020-Septe 1–4 (2020).
    1. Demonbreun, A. & Mirsky, G. M. Automated classification of electrocardiograms using wavelet analysis and deep learning. In Computing in Cardiology Vol. 2020-Septe 12–15 (2020).
    1. Miwakeichi F, et al. Decomposing EEG data into space-time-frequency components using Parallel Factor Analysis. Neuroimage. 2004;22:1035–1045. doi: 10.1016/j.neuroimage.2004.03.039.
    1. Hurnanen T, et al. Automated detection of atrial fibrillation based on time-frequency analysis of seismocardiograms. IEEE J. Biomed. Heal. Inform. 2017;21:1233–1241. doi: 10.1109/JBHI.2016.2621887.
    1. Stridh M, Sörnmo L, Meurling CJ, Olsson SB. Sequential characterization of atrial tachyarrhythmias based on ECG time-frequency analysis. IEEE Trans. Biomed. Eng. 2004;51:100–114. doi: 10.1109/TBME.2003.820331.
    1. Chun, S. Y. et al. ECG based user authentication for wearable devices using short time Fourier transform. In 2016 39th Int. Conf. Telecommun. Signal Process. TSP 2016 656–659 10.1109/TSP.2016.7760964 (2016).
    1. Ng J, Goldberger JJ. Understanding and interpreting dominant frequency analysis of AF electrograms. J. Cardiovasc. Electrophysiol. 2007;18:680–685. doi: 10.1111/j.1540-8167.2007.00832.x.
    1. Benmalek, E. & Elmhamdi, J. Arrhythmia ECG signal analysis using non parametric time-frequency technique. In Proc. 2015 Int. Conf. Electr. Inf. Technol. ICEIT 2015 281–285 10.1109/EITech.2015.7162958 (2015).
    1. Classification of 12-lead ECGs: the PhysioNet/computing in cardiology challenge 2020. In Computers in Cardiology (2020).
    1. Jeni, L. A., Cohn, J. F. & De La Torre, F. Facing imbalanced data—Recommendations for the use of performance metrics. In Proc.—2013 Hum. Assoc. Conf. Affect. Comput. Intell. Interact. ACII 2013 245–251 10.1109/ACII.2013.47 (2013).
    1. Fawcett T. An introduction to ROC analysis. Pattern Recogn. Lett. 2006;27:861–874. doi: 10.1016/j.patrec.2005.10.010.
    1. Liu, F. F. et al. The China physiological signal challenge 2018: automatic identification of the rhythm/morphology abnormalities in 12-lead ECGs. In The 7th International Conference on Biomedical Engineering and Biotechnology (ICBEB 2018) (2018).
    1. Liu F, et al. An open access database for evaluating the algorithms of electrocardiogram rhythm and morphology abnormality detection. J. Med. Imaging Health Inform. 2018;8:1368–1373. doi: 10.1166/jmihi.2018.2442.
    1. Pan J, Tompkins WJ. A real-time QRS detection algorithm. IEEE Trans. Biomed. Eng. 1985;BME-32:230–236. doi: 10.1109/TBME.1985.325532.
    1. Kathirvel P, Manikandan MS, Prasanna SRM, Soman KP. An efficient R-peak detection based on new nonlinear transformation and first-order gaussian differentiator. Cardiovasc. Eng. Technol. 2011;2:408–425. doi: 10.1007/s13239-011-0065-3.
    1. Lloyd SP. Least squares quantization in PCM. IEEE Trans. Inf. Theory. 1982;28:129–137. doi: 10.1109/TIT.1982.1056489.
    1. Arthur D, Vassilvitskii S. k-means++: The Advantages of Careful Seeding. Stanford; 2006.
    1. Elkan, C. Using the triangle inequality to accelerate k-means. In Proceedings, Twent. Int. Conf. Mach. Learn. Vol. 1, 147–153 (2003).
    1. Tereshchenko LG, Josephson ME. Frequency content and characteristics of ventricular conduction. J. Electrocardiol. 2015;48:933–937. doi: 10.1016/j.jelectrocard.2015.08.034.
    1. Xavier, G., Antoine, B. & Yoshua, B. Deep sparse rectifier neural networks. In International Conference on Artificial Intelligence and statistics (AISTATS) Vol. 15, 315–323 (2011).
    1. Kingma, D. P. & Ba, J. L. Adam: a method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015—Conference Track Proceedings 1–15 (2015).

Source: PubMed

Sponsorit ja yhteistyökumppanit

Sairaudet

Huumeiden interventiot

3
Tilaa