DOI QR코드

DOI QR Code

Predicton and Elapsed time of ECG Signal Using Digital FIR Filter and Deep Learning

디지털 FIR 필터와 Deep Learning을 이용한 ECG 신호 예측 및 경과시간

  • Received : 2023.05.27
  • Accepted : 2023.07.03
  • Published : 2023.07.31

Abstract

ECG(electrocardiogram) is used to measure the rate and regularity of heartbeats, as well as the size and position of the chambers, the presence of any damage to the heart, and the cause of all heart diseases can be found. Because the ECG signal obtained using the ECG-KIT includes noise in the ECG signal, noise must be removed from the ECG signal to apply to the deep learning. In this paper, Noise included in the ECG signal was removed by using a lowpass filter of the Digital FIR Hamming window function. When the performance evaluation of the three activation functions, sigmoid(), ReLU(), and tanh() functions, which was confirmed that the activation function with the smallest error was the tanh() function, the elapsed time was longer when the batch size was small than large. Also, it was confirmed that result of the performance evaluation for the GRU model was superior to that of the LSTM model.

심전도(electrocardiogram, ECG)는 심박동의 속도와 규칙성, 심실의 크기와 위치, 심장 손상 여부를 측정하는 데 사용되며, 모든 심장질환의 원인을 찾아낼 수 있다. ECG-KIT를 이용하여 획득한 ECG 신호는 ECG 신호에 잡음을 포함하기 때문에 딥러닝에 적용하기 위해서는 ECG 신호에서 잡음을 제거해야만 한다. 본 논문에서는, ECG 신호에 포함된 잡음은 Digital FIR 해밍 창함수를 이용한 저역통과 필터를 사용하여 제거하였다. LSTM의 딥러닝 모델을 사용하여 3가지 활성화 함수인 sigmoid(), ReLU(), tanh() 에 대한 성능 평가를 비교했을 때, 오차가 가장 작은 활성화 함수는 tanh() 함수 임을 확인하였고, batch size가 작은 경우가 큰 경우보다 시간이 많이 소요되었다. 또한 GRU 모델의 성능 평가의 결과가 LSTM 모델보다 우수한 것을 확인하였다.

Keywords

References

  1. 최윤식, 임상심전도학(제5판), 교보문고. 2009 
  2. https://a-fib.com/treatments-for-atrial-fibrillation/diagnostic-tests-2/the-ekg-signal/ 
  3. https://physiolab.en.ec21.com/ECG-KIT Assembly-ECG-sensor-9570299_9570369.html 
  4. 하석운, 파이썬으로 배우는 디지털 신호처리, 생능 출판사, 2020. 
  5. Aurelien Geron, Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow(2nd ed.), O'Reilly Media, Inc, 2021. 
  6. S. Hochreiter and J. Schmidhuber, "Long shortterm memory," Neural Computation, vol. 9, no. 8, pp. 1735-1780, Nov. 1997. DOI: https://doi.org/10.1162/neco.1997.9.8.1735 
  7. Kyunghyun Cho et al., "Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation," Procedings of the 2014 Conference on Empirical Methods in Natural Language Processing (2014): pp. 1724-1734, Sep. 2014. https://doi.org/10.48550/arXiv.1406.1078 
  8. Kun-Woo Kim, et al, "Anomaly Detection of Air Dryer for Radar based on Machine Learning Algorithms," Journal of the Korea AcademiaIndustrial cooperation Society(JKAIS), Vol. 24, No. 3 pp. 1-12, 2023.  https://doi.org/10.5762/KAIS.2023.24.3.1
  9. Jin-Young Kim et al, "Prediction System of Running Heart Rate based on FitRec," The Journal of the Institute of Internet, Broadcasting and Communication(JIIBC), Vol. 22, No. 6, pp. 165-171, Dec. 31, 2022. https://doi.org/10.7236/JIIBC.2022.22.6.165 
  10. J. H. Kim et al, "Prediction of Battery Remaining Life based on Sliding Window-LSTM for Camouflage Therapeutic Syste," The Journal of Journal of KIIT(JKIIT), Vol. 21, No. 1, pp. 29-38, Jan. 31, 2023. http://dx.doi.org/10.14801/jkiit.2023.21.1.29