Skip to main content
SpringerLink
Log in

Deep convolutional neural networks based ECG beats classification to diagnose cardiovascular conditions

  • Original Article
  • Published:
Biomedical Engineering Letters Aims and scope Submit manuscript

Abstract

Medical practitioners need to understand the critical features of ECG beats to diagnose and identify cardiovascular conditions accurately. This would be greatly facilitated by identifying the significant features of frequency components in temporal ECG wave-forms using computational methods. In this study, we have proposed a novel ECG beat classifier based on a customized VGG16-based Convolution Neural Network (CNN) that uses the time-frequency representation of temporal ECG, and a method to identify the contribution of interpretable ECG frequencies when classifying based on the SHapley Additive exPlanations (SHAP) values. We applied our model to the MIT-BIH arrhythmia dataset to classify the ECG beats and to characterise of the beats frequencies. This model was evaluated with two advanced time-frequency analysis methods. Our results indicated that for 2-4 classes our proposed model achieves a classification accuracy of 100% and for 5 classes it achieves a classification accuracy of 99.90%. We have also tested the proposed model using premature ventricular contraction beats from the American Heart Association (AHA) database and normal beats from Lobachevsky University Electrocardiography database (LUDB) and obtained a classification accuracy of 99.91% for the 5-classes case. In addition, SHAP value increased the interpretability of the ECG frequency features. Thus, this model could be applicable to the automation of the cardiovascular diagnosis system and could be used by clinicians.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

References

  1. Benjamin EJ, Muntner P, Sommer MB. Heart disease and stroke statistics-2019 update: a report from the American heart association. Circulation. 2019;139(10):e56–528.

    Article  Google Scholar 

  2. Narain SY, Kumar SSS, Kumar RA. Bioelectrical signals as emerging biometrics: issues and challenges; 2012. p. 2012.

  3. Mohamed EAB, Dong GL, Makki AA, Gyeong MY, Eun-jong C, Jang-whan B, Myeong CC, Keun HR. Highlighting the current issues with pride suggestions for improving the performance of real time cardiac health monitoring. In: International Conference on Information Technology in Bio-and Medical Informatics, pp 226–233. Springer, 2010.

  4. Shweta HJ, Vipul KD, Harshadkumar BP. Classification of ecg signals using machine learning techniques: a survey. In: 2015 International Conference on Advances in Computer Engineering and Applications, pp. 714–721. IEEE, 2015.

  5. Rajendra AU, Paul JK, Kannathal N, Lim CM, Suri JS. Heart rate variability: a review. Med Biol Eng Comput. 2006;44(12):1031–51.

    Article  Google Scholar 

  6. Marwa MAH, Mohamed IE, Ahmed F. Computer aided diagnosis of cardiac arrhythmias. In: 2006 International Conference on Computer Engineering and Systems, pp. 262–265. IEEE, 2006.

  7. De Chazal P, O’Dwyer M, Reilly RB. Automatic classification of heartbeats using ecg morphology and heartbeat interval features. IEEE Trans Biomed Eng. 2004;51(7):1196–206.

    Article  Google Scholar 

  8. Terrill F, David HW. A minicomputer system for direct high speed analysis of cardiac arrhythmia in 24 h ambulatory ecg tape recordings. IEEE Trans Biomed Eng. 1980;12:685–93.

    Google Scholar 

  9. Coast Douglas A, Stern Richard M, Cano Gerald G, Briller SA. An approach to cardiac arrhythmia analysis using hidden markov models. IEEE Trans Biomed Eng. 1990;37(9):826–36.

    Article  Google Scholar 

  10. Lin K-P, Chang WH. Qrs feature extraction using linear prediction. IEEE Trans Biomed Eng. 1989;36(10):1050–5.

    Article  Google Scholar 

  11. Farid M, Yakoub B. Classification of electrocardiogram signals with support vector machines and particle swarm optimization. IEEE Trans Inf Technol Biomed. 2008;12(5):667–77.

    Article  Google Scholar 

  12. Ge D, Srinivasan N, Krishnan SM. Cardiac arrhythmia classification using autoregressive modeling. Biomed Eng. 2002;1(1):5.

    Google Scholar 

  13. Mishra AK, Raghav S. Local fractal dimension based ecg arrhythmia classification. Biomed Signal Process Control. 2010;5(2):114–23.

    Article  Google Scholar 

  14. Ali K, Ataollah E. Classification of electrocardiogram signals with support vector machines and genetic algorithms using power spectral features. Biomed Signal Process Control. 2010;5(4):252–63.

    Article  Google Scholar 

  15. Linh TH, Osowski S, Stodolski M. On-line heart beat recognition using hermite polynomials and neuro-fuzzy network. IEEE Trans Instrum Meas. 2003;52(4):1224–311.

    Article  Google Scholar 

  16. Martis RJ, Chakraborty C, Ray AK. A two-stage mechanism for registration and classification of ecg using gaussian mixture model. Pattern Recognit. 2009;42(11):2979–88.

    Article  MATH  Google Scholar 

  17. Abdelhamid D, Latifa H, Naif A, Farid M. A wavelet optimization approach for ecg signal classification. Biomed Signal Process Control. 2012;7(4):342–9.

    Article  Google Scholar 

  18. Martis RJ, Muthu RKM, Chakraborty C, Pal S, Sarkar D, Mandana KM, Ray AK. Automated screening of arrhythmia using wavelet based machine learning techniques. J Med Syst. 2012;36(2):677–88.

    Article  Google Scholar 

  19. Hamid K, Majid M. A comparative study of dwt, cwt and dct transformations in ecg arrhythmias classification. Expert Syst Appl. 2010;37(8):5751–7.

    Article  Google Scholar 

  20. Yu HH, Palreddy S, Tompkins WJ. A patient-adaptable ecg beat classifier using a mixture of experts approach. IEEE Trans Biomed Eng. 1997;44(9):891–900.

    Article  Google Scholar 

  21. Jiang W, Kong SG. Block-based neural networks for personalized ecg signal classification. IEEE Trans Neural Netw. 2007;18(6):1750–61.

    Article  Google Scholar 

  22. 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(12):2507–15.

    Article  Google Scholar 

  23. Martis RJ, Rajendra AU, Mandana KM, Ray AK, Chakraborty C. Application of principal component analysis to ecg signals for automated diagnosis of cardiac health. Expert Syst Appl. 2012;39(14):11792–800.

    Article  Google Scholar 

  24. Martis RJ, Rajendra AU, Mandana KM, Ray AK, Chakraborty C. Cardiac decision making using higher order spectra. Biomed Signal Process Control. 2013;8(2):193–203.

    Article  Google Scholar 

  25. Rajendra AU, Fujita H, Shu LO, Raghavendra U, Tan JH, Adam M, Gertych Arkadiusz HY. Automated identification of shockable and non-shockable life-threatening ventricular arrhythmias using convolutional neural network. Fut Gen Comput Syst. 2018;79:952–9.

    Article  Google Scholar 

  26. De Chazal P, Reilly RB. A patient-adapting heartbeat classifier using ecg morphology and heartbeat interval features. IEEE Trans Biomed Eng. 2006;53(12):2535–43.

    Article  Google Scholar 

  27. Sannino G, De Pietro G. A deep learning approach for ecg-based heartbeat classification for arrhythmia detection. Fut Gen Comput Syst. 2018;86:446–55.

    Article  Google Scholar 

  28. Mathews SM, Kambhamettu C, Barner KE. A novel application of deep learning for single-lead ecg classification. Comput Biol Med. 2018;99:53–62.

    Article  Google Scholar 

  29. Rajendra AU, Hamido FO, Lih S, Hagiwara Y, Tan JH, Adam M. Automated detection of arrhythmias using different intervals of tachycardia ecg segments with convolutional neural network. Inf Sci. 2017;405:81–90.

    Article  Google Scholar 

  30. Jambukia Shweta H, Dabhi Vipul K, Prajapati HB. Ecg beat classification using machine learning techniques. Int J Biomed Eng Technol. 2018;26(1):32–533.

    Article  Google Scholar 

  31. Rajendra AU, Shu LO, Hagiwara Y, Tan JH, Adam M, Gertych A, Tan RS. A deep convolutional neural network model to classify heartbeats. Comput Biol Med. 2017;89:389–96.

    Article  Google Scholar 

  32. Rajendra AU, Fujita H, Shu LO, Hagiwara Y, Tan JH, Adam M. Application of deep convolutional neural network for automated detection of myocardial infarction using ecg signals. Inf Sci. 2017;415:190–8.

    Google Scholar 

  33. Ali I, Selen O. Cardiac arrhythmia detection using deep learning. Procedia Comput Sci. 2017;120:268–75.

    Article  Google Scholar 

  34. Rajendra AU, Fujita H, Shu LO, Hagiwara Y, Tan JH, Adam M, Tan RS. Deep convolutional neural network for the automated diagnosis of congestive heart failure using ecg signals. Appl Intell. 2019;49(1):16–27.

    Article  Google Scholar 

  35. Karen S, Andrew Z. Very deep convolutional networks for large-scale image recognition. 2014. arXiv preprint arXiv:1409.1556

  36. Moody George B, Mark RG. The impact of the mit-bih arrhythmia database. IEEE Eng Med Biol Mag. 2001;20(3):45–50.

    Article  Google Scholar 

  37. R Mark, G Moody. Mit-bih database and software catalog, 1997.

  38. Ziqian W, Xujian F, Cuiwei Y. A deep learning method to detect atrial fibrillation based on continuous wavelet transform. In: 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pp. 1908–1912. IEEE, 2019.

  39. Norden EH, Zheng S, Steven RL, Manli CW, Hsing HS, Quanan Z, Nai-Chyuan Y, Chi CT, Henry HL. The empirical mode decomposition and the hilbert spectrum for nonlinear and non-stationary time series analysis. Proc R Soc Lond Ser A Math Phys Eng Sci. 1998;454(1971):903–95.

    Article  MathSciNet  MATH  Google Scholar 

  40. Scott M L, Su-In L. A unified approach to interpreting model predictions. In I. Guyon, U. V. Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan, R. Garnett, editors, Advances in Neural Information Processing Systems 30, pp. 4765–4774. Curran Associates, Inc., 2017.

  41. Gabriel E, Joseph DJ, Pascal S, Scott L, Su-In L. Learning explainable models using attribution priors. 2019. arXiv preprint arXiv:1906.10670

  42. Mukund S, Ankur T, Qiqi Y. Axiomatic attribution for deep networks. In: Proceedings of the 34th International Conference on Machine Learning-Volume 70, pp. 3319–3328. JMLR. org, 2017.

  43. François C et al. Keras. https://github.com/fchollet/keras, 2015.

  44. Ary LG, Luis ANA, Leon G, Jeffrey MH, Plamen CI, Roger GM, Joseph EM, George BM, Chung-Kang P, Stanley HE. Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals. Circulation. 2000;101(23):e215–20.

    Google Scholar 

  45. Kalyakulina A, Yusipov II, Moskalenko VA, Nikolskiy AV, Kozlov AA, Kosonogov KA, Zolotykh NY, Ivanchenko MV. Lobachevsky university electrocardiography database (version 1.0. 0). PhysioNet, 2020.

  46. Rai HM, Chatterjee K. A novel adaptive feature extraction for detection of cardiac arrhythmias using hybrid technique mrdwt & mpnn classifier from ecg big data. Big Data Res. 2018;12:13–22.

    Article  Google Scholar 

  47. Güler İnan, Übeylı Elif D. Ecg beat classifier designed by combined neural network model. Pattern Recognit. 2005;38(2):199–208.

  48. Bahareh T, Adrian DCC, Shervin S. Classifying measured electrocardiogram signal quality using deep belief networks. In: 2017 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), pp. 1–6. IEEE, 2017.

  49. Serkan K, Turker I, Moncef G. Real-time patient-specific ecg classification by 1-d convolutional neural networks. IEEE Trans Biomed Eng. 2015;63(3):664–75.

    Google Scholar 

  50. Shu LO, Ng EYK, Tan RS, Rajendra AU. Automated diagnosis of arrhythmia using combination of cnn and lstm techniques with variable length heart beats. Comput Biol Med. 2018;102:278–87.

    Article  Google Scholar 

  51. Gramatikov B, Brinker J, Yi-Chun S, Thakor NV. Wavelet analysis and time-frequency distributions of the body surface ecg before and after angioplasty. Comput Methods Progr Biomed. 2000;62(2):87–988.

    Article  Google Scholar 

  52. Takano NK, Tsutsumi T, Suzuki H, Okamoto Y, Nakajima T. Time frequency power profile of qrs complex obtained with wavelet transform in spontaneously hypertensive rats. Comput Biol Med. 2012;42(2):205–12.

    Article  Google Scholar 

  53. Takeshi T, Yoshiwo O, Nami K-T, Daisuke W, Hiroshi S, Kazunori S, Kuniaki I, Toshiaki N. Time-frequency analysis of the qrs complex in patients with ischemic cardiomyopathy and myocardial infarction. IJC Heart Ves. 2014;4:177–87.

    Article  Google Scholar 

  54. Ernest W, Reynolds JR, Muller B, Anderson GJ, Muller BT. High-frequency components in the electrocardiogram: a comparative study of normals and patients with myocardial disease. Circulation. 1967;35(1):195–206.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Computer Science and Engineering, University of Rajshahi, Rajshahi, 6205, Bangladesh

    Md. Rashed-Al-Mahfuz

  2. WHO Collaborating Centre of eHealth, School of Public Health and Community Medicine, The University of New South Wales, Kensington, NSW, 2052, Australia

    Mohammad Ali Moni

  3. Computer Laboratory, The University of Cambridge, 15 JJ Thomson Avenue, Cambridge, UK

    Pietro Lio’

  4. Faculty of Health, School of Exercise and Nutrition Sciences′ to ′Institute for Physical Activity and Nutrition, School of Exercise and Nutrition Sciences, Deakin University, Geelong, Australia

    Sheikh Mohammed Shariful Islam

  5. Australian Institute of Health Innovation, Faculty of Medicine and Health Sciences, Macquarie University, Sydney, Australia

    Shlomo Berkovsky

  6. School of Computer Science, The University of Sydney, Sydney, NSW, 2006, Australia

    Matloob Khushi

  7. Healthy Ageing Theme, Garvan Institute of Medical Research, Darlinghurst, NSW, 2010, Australia

    Mohammad Ali Moni & Julian M. W. Quinn

  8. Royal North Shore Hospital, SERT Institute, St. Leonards, NSW, Australia

    Julian M. W. Quinn

Authors
  1. Md. Rashed-Al-Mahfuz
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Mohammad Ali Moni
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Pietro Lio’
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Sheikh Mohammed Shariful Islam
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Shlomo Berkovsky
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Matloob Khushi
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Julian M. W. Quinn
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Mohammad Ali Moni.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

We have used publicly available data in our experiment. Therefore we do not need any ethical consent for this study.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Rashed-Al-Mahfuz, M., Moni, M.A., Lio’, P. et al. Deep convolutional neural networks based ECG beats classification to diagnose cardiovascular conditions. Biomed. Eng. Lett. 11, 147–162 (2021). https://doi.org/10.1007/s13534-021-00185-w

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s13534-021-00185-w

Keywords

Search

Navigation