Electrocardiogram (ECG) is strong evidence in the diagnosis of a wide range of heart-related diseases, and it is becoming increasingly important in the medical field recently. However, inferencing diseases with ECG signals is both time-consuming and error-prone even for licensed physicians, which arises the urgency of developing a fast and accurate automatic diagnosis algorithm. In this paper, we explore both deep learning models and well-designed feature engineering from ECG waveform. By combining the two methods, we propose an automatic diagnosis framework that can extract meaningful features both with and without human interventions. Experimental results on the ECG competition demonstrate that our framework can reach accurate results on heart-related diseases diagnosis.


ECG Deep learning Feature engineering Automatic diagnosis framework 



The work was supported by the National Natural science Foundation of China (NSFC) Projects (Nos. 61673241, 61721003, 61872218), Beijing National Research Center for Information Science and Technology, Tsinghua-Fuzhou Institute research program, and Tsinghua Institute of Data Sciences.


  1. 1.
    The first China ECG intelligent competition. Accessed 25 June 2019
  2. 2.
    Acharya, U.R., et al.: A deep convolutional neural network model to classify heartbeats. Comput. Biol. Med. 89, 389–396 (2017)CrossRefGoogle Scholar
  3. 3.
    Chauhan, S., Vig, L.: Anomaly detection in ECG time signals via deep long short-term memory networks, pp. 1–7 (2015)Google Scholar
  4. 4.
    Cho, K., et al.: Learning phrase representations using RNN encoder-decoder for statistical machine translation. arXiv preprint arXiv:1406.1078 (2014)
  5. 5.
    Daamouche, A., Hamami, L., Alajlan, N., Melgani, F.: A wavelet optimization approach for ECG signal classification. Biomed. Signal Process. Control 7(4), 342–349 (2012) CrossRefGoogle Scholar
  6. 6.
    Feng, Y., Chen, W., Cai, G.: Feature extraction and identification of biometric information from ECG. Comput. Digit. Eng. 46(6), 1099–1103 (2018)Google Scholar
  7. 7.
    Hannun, A., et al.: Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nat. Med. 25(1), 65–69 (2019)CrossRefGoogle Scholar
  8. 8.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  9. 9.
    Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization. In: Proceedings of the Third International Conference on Learning Representations (2015)Google Scholar
  10. 10.
    Mikolov, T., Karafiát, M., Burget, L., Černockỳ, J., Khudanpur, S.: Recurrent neural network based language model. In: Eleventh Annual Conference of the International Speech Communication Association (2010)Google Scholar
  11. 11.
    Pan, G., Xin, Z., Shi, S., Jin, D.: Arrhythmia classification based on wavelet transformation and random forests. Multimedia Tools Appl. 77(17), 21905–21922 (2018)CrossRefGoogle Scholar
  12. 12.
    Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., Salakhutdinov, R.: Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15(1), 1929–1958 (2014)MathSciNetzbMATHGoogle Scholar
  13. 13.
    Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems, pp. 5998–6008 (2017)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Tsinghua UniversityBeijingChina

Personalised recommendations