Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Arrhythmia Classification using Hybrid Combination Model of CNN-LSTM

합성곱-장단기 기억 신경망의 하이브리드 결합 모델을 이용한 부정맥 분류

  • Cho, Ik-Sung (School of Interdisciplinary Studies, Daegu University,) ;
  • Kwon, Hyeog-Soong (Department of IT Engineering, Pusan National University)
  • Received : 2021.10.10
  • Accepted : 2021.10.26
  • Published : 2022.01.31
Download PDF
⟨ Previous Next ⟩

Abstract

Arrhythmia is a condition in which the heart beats abnormally or irregularly, early detection is very important because it can cause dangerous situations such as fainting or sudden cardiac death. However, performance degradation occurs due to personalized differences in ECG signals. In this paper, we propose arrhythmia classification using hybrid combination model of CNN-LSTM. For this purpose, the R wave is detected from noise removed signal and a single bit segment was extracted. It consisted of eight convolutional layers to extract the features of the arrhythmia in detail, used them as the input of the LSTM. The weights were learned through deep learning and the model was evaluated by the verification data. The performance was compared in terms of the accuracy, precision, recall, F1 score through MIT-BIH arrhythmia database. The achieved scores indicate 92.3%, 90.98%, 92.20%, 90.72% in terms of the accuracy, precision, recall, F1 score, respectively.

부정맥은 심장 박동이 비정상 혹은 불규칙하게 뛰고 있는 상태를 말하며, 실신이나 심장돌연사 등과 같은 위험한 상황을 유발할 수 있기 때문에 이의 조기 검출은 매우 중요하다. 하지만 심전도 신호의 개인차로 인해 분류 시 성능하락이 나타날 수밖에 없다. 본 연구에서는 CNN-LSTM 하이브리드 결합 모델을 이용한 부정맥 분류 방법을 제안한다. 이를 위해 먼저 잡음을 제거한 ECG 신호에서 R파를 검출하고 단일 비트 세그먼트를 추출하였다. 이후 부정맥 신호의 특징을 세밀하게 추출하도록 8개의 합성곱 계층으로 구성하고 이를 LSTM의 입력으로 사용한 후 가중치를 학습시키고 검증 데이터로 모델을 평가한 후 정상 및 부정맥 분류의 변화를 확인하였다. 제안한 방법의 타당성 검증을 위해 MIT-BIH 부정맥 데이터베이스를 사용하여 정확도(accuracy), 정밀도(precision), 재현율(recall), F1 스코어가 사용되었다. 성능평가 결과, 정확도, 정밀도, 재현율, F1 스코어는 각각 92.3%, 90.98%, 92.20%, 90.72%의 우수한 분류율을 나타내었다.

Keywords

Acknowledgement

This work was supported by a 2-Year Research Grant of Pusan National University

References

  1. H. M. Tun, W. K. Moe, and Z. M. Naing, "Analysis on conversion process from paper record ECG to computer based ECG," MedCrave Online Journal of Applied Bionics and Biomechanics, vol. 1, no. 2, pp. 69-81, Sep. 2017.
  2. A. Hossain, R. Quaresma, and H. Rahman, "Investigating factors influencing the physicians' adoption of electronic health record (EHR) in healthcare system of Bangladesh: An empirical study," International Journal of Information Management, vol. 44, pp. 76-87, Feb. 2019. https://doi.org/10.1016/j.ijinfomgt.2018.09.016
  3. Q. Qin, J. Li, Y. Yue, and C. Liu, "An Adaptive and Time-Efficient ECG R-Peak Detection Algorithm," Journal of Healthcare Engineering, vol. 2017, pp. 1-14, Sep. 2017.
  4. M. J. Goldman, Principles of Clinical Electrocardiography, Los Altos California: Lange Medical Pubns, 1986.
  5. K. Hanbay, "Deep neural network based approach for ECG classification using hybrid differential features and active learning," Institution of Engineering and Technology, vol. 13, no. 2, pp. 165-175, May. 2019.
  6. W. Li, "Deep Intermediate Representation and In-Set Voting Scheme for Multiple-Beat Electrocardiogram Classification," IEEE Sensors Journal, vol. 19, no. 16, pp. 6895-6904, Apr. 2019. https://doi.org/10.1109/jsen.2019.2910853
  7. M. Altuve and F. Hernandez, "Multiclass Classification of Cardiac Rhythms on Short Single Lead ECG Recordings using Bidirectional Long Short-Term Memory Networks," IEEE Latin America Transactions, vol. 19, no. 7, pp. 1207-1216, Jul. 2021. https://doi.org/10.1109/TLA.2021.9461850
  8. I. S. Cho and H. S. Kwon, "Optimal Threshold Setting Method for R Wave Detection According to The Sampling Frequency of ECG Signals," Journal of Korea Institute of Information and Communication Engineering, vol. 21, no. 7, pp. 1420-1428, Jul. 2017. https://doi.org/10.6109/JKIICE.2017.21.7.1420
  9. Y. Wei, J. Zhou, Y. Wang, Y. Liu, Q. Liu, J. Luo, C. Wang, F Ren, and L. Huang, "A Review of Algorithm & Hardware Design for AI-Based Biomedical Applications," IEEE Transactions on Biomedical Circuits and Systems, vol. 14, no. 2, pp. 145-163, Apr. 2020. https://doi.org/10.1109/tbcas.2020.2974154
  10. S. Xu, J. Li, K. Liu, and L. Wu, "A Parallel GRU Recurrent Network Model and its Application to Multi-Channel Time-Varying Signal Classification," IEEE Access, vol. 7, pp. 118739-118748, Aug. 2019. https://doi.org/10.1109/access.2019.2936516
  11. Y. Li, Z. Xia, and Y. Zhang, "Standalone Systolic Profile Detection of Non-Contact SCG Signal With LSTM Network," IEEE Sensors Journal, vol. 20, no. 6, pp. 3123-3131, Mar. 2020. https://doi.org/10.1109/jsen.2019.2957382
  12. R. He, Y. Liu, K. Wang, N. Zhao, Y. Yuan, Q. Li, and H. Zhang, "Automatic Cardiac Arrhythmia Classification Using Combination of Deep Residual Network and Bidirectional LSTM," IEEE Access, vol. 7, pp. 102119-102135, Jul. 2019. https://doi.org/10.1109/access.2019.2931500
  13. S. Singh, S. K. Pandey, U. Pawar, and R. R. Janghel, "Classification of ECG Arrhythmia using Recurrent Neural Networks," Procedia Computer Science, vol. 132, pp. 1290-1297, Jun. 2018. https://doi.org/10.1016/j.procs.2018.05.045
  14. A. Sellami and H. Hwang, "A robust deep convolutional neural network with batch-weighted loss for heartbeat classification," Expert Systems with Applications, vol. 122, pp. 75-84, May. 2019. https://doi.org/10.1016/j.eswa.2018.12.037