Multi-label Classification of Abnormalities in 12-Lead ECG Using 1D CNN and LSTM

  • Chengsi Luo
  • Hongxiu Jiang
  • Quanchi Li
  • Nini RaoEmail author
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11794)


In this study, we proposed a method based on Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to classify 12-lead ECG into 9 categories (1 normal, 8 abnormal). The only preprocessing techniques we used are the baseline drift removal based on median filtering and signal segmentation. Then an 18-layer deep 1D CNN consisting of residual blocks and skip architectures that is followed by a bi-directional LSTM layer was developed. During the training session, we suggested a new strategy to compute the Dice loss for multi-label classification. The average F1-score we achieved on the hidden testing dataset of the First China ECG Intelligent Competition (FCEIC) is 85.11%. With the same model, we achieved 82.21% (5-fold cross-validation) on the Chinese physiological signal challenge 2018 (CPSC2018) training dataset.


ECG classification Convolutional Neural Network Dice loss Long Short-Term Memory 



This work was supported by National Natural Science Foundation of China (Grant No. 61872405 and 61720106004), Key Project of Natural Science Foundation of Guangdong province (2016A030311040), Sichuan Science and Support Program (Grant No. 2015SZ0191) and Chengdu Science and Technology Benefit Plan (2015-HM01-00528-SF).


  1. 1.
    World Health Organization: World health statistics 2018: Monitoring Health for the SDGs sustainable development goals. WHO (2018)Google Scholar
  2. 2.
    Bizopoulos, P., Koutsouris, D.: Deep learning in cardiology. IEEE Rev. Biomed. Eng. 12, 168–193 (2019). Scholar
  3. 3.
    Gotlibovych, I., et al.: End-to-end deep learning from raw sensor data: atrial fibrillation detection using wearables. arXiv preprint arXiv:1807.10707 (2018)
  4. 4.
    Zihlmann, M., Perekrestenko, D., Tschannen, M.: Convolutional recurrent neural networks for electrocardiogram classification. In: 2017 Computing in Cardiology Conference (CinC) (2017)Google Scholar
  5. 5.
    Bashivan, P., Rish, I., Yeasin, M., Codella, N.: Learning representations from EEG with deep recurrent-convolutional neural networks. arXiv preprint arXiv:1511.06448 (2015)
  6. 6.
    Sellami, A., Hwang, H.: A robust deep convolutional neural network with batch-weighted loss for heartbeat classification. Expert Syst. Appl. 122, 75–84 (2019). Scholar
  7. 7.
    Liu, Z., Meng, X.A., Cui, J., Huang, Z., Wu, J.: Automatic identification of abnormalities in 12-lead ECGs using expert features and convolutional neural networks. In: 2018 International Conference on Sensor Networks and Signal Processing (SNSP), pp. 163–167. IEEE (2018)Google Scholar
  8. 8.
    Hannun, A.Y., et al.: Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nat. Med. 25, 65–69 (2019). Scholar
  9. 9.
    The First China ECG Intelligent Competition (2019).
  10. 10.
    Liu, F.F., et al.: An open access database for evaluating the algorithms of electrocardiogram rhythm and morphology abnormality detection. J. Med. Imaging Health Inform. 8, 1368–1373 (2018). Scholar
  11. 11.
    The China Physiological Signal Challenge (2018).
  12. 12.
    Lenis, G., Pilia, N., Loewe, A., Schulze, W.H., Dossel, O.: Comparison of baseline wander removal techniques considering the preservation of ST changes in the ischemic ECG: a simulation study. Comput. Math. Methods Med. 2017, 9295029 (2017)CrossRefGoogle Scholar
  13. 13.
    Clevert, D.-A., Unterthiner, T., Hochreiter, S.: Fast and accurate deep network learning by exponential linear units (ELUs). arXiv preprint arXiv:1511.07289 (2015)
  14. 14.
    Chollet, F.: Xception: deep learning with depthwise separable convolutions. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1800–1807 (2017)Google Scholar
  15. 15.
    Edenbrandt, L., Pahlm, O.: Vectorcardiogram synthesized from a 12-lead ECG: superiority of the inverse Dower matrix. J. Electrocardiol. 21, 361–367 (1988)CrossRefGoogle Scholar
  16. 16.
    Long, J., Shelhamer, E., Darrell, T.: Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 3431–3440 (2015)Google Scholar
  17. 17.
    Saeedan, F., Weber, N., Goesele, M., Roth, S.: Detail-preserving pooling in deep networks. In: 2018 Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 9108–9116 (2018)Google Scholar
  18. 18.
    Tan, J.H., et al.: Application of stacked convolutional and long short-term memory network for accurate identification of CAD ECG signals. Comput. Biol. Med. 94, 19–26 (2018). Scholar
  19. 19.
    Dyer, C., Ballesteros, M., Ling, W., Matthews, A., Smith, N.A.: Transition-based dependency parsing with stack long short-term memory. arXiv preprint arXiv:1505.08075 (2015)
  20. 20.
    Zhu, W., et al.: AnatomyNet: deep learning for fast and fully automated whole-volume segmentation of head and neck anatomy. Med. Phys. 46, 576–589 (2019). Scholar
  21. 21.
    Wong, K.C.L., Moradi, M., Tang, H., Syeda-Mahmood, T.: 3D segmentation with exponential logarithmic loss for highly unbalanced object sizes. In: Frangi, A.F., Schnabel, J.A., Davatzikos, C., Alberola-López, C., Fichtinger, G. (eds.) MICCAI 2018. LNCS, vol. 11072, pp. 612–619. Springer, Cham (2018). Scholar
  22. 22.
    Chollet, F., et al.: Keras (2015).
  23. 23.
    Zhang, M.L., Zhou, Z.H.: A review on multi-label learning algorithms. IEEE Trans. Knowl. Data Eng. 26, 1819–1837 (2014)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Chengsi Luo
    • 1
  • Hongxiu Jiang
    • 1
  • Quanchi Li
    • 1
  • Nini Rao
    • 1
    Email author
  1. 1.School of Life Science and TechnologyUniversity of Electronic Science and Technology of ChinaChengduPeople’s Republic of China

Personalised recommendations