心臟疾病在國際上是死亡的主要原因之一，而心電訊號的監測更為判斷是否有患病可能的重要前提。心電訊號中最明顯的波型是R波，因此R波的偵測與判讀是分析心電訊號的第一步驟。我們提出了時域多層級R波偵測演算法，此方法不僅複雜度低，也可以維持波形的完整性，對於後續的疾病偵測有極大的幫助。本研究對其中的係數做最佳化的處理，R波偵測演算法本質上也是個二元分類器，因此在最佳化係數的過程當中，將以分類器性能評估的圖形方法作為主要的工具。本研究利用PhysioNet所提供之MIT-BIH arrhythmia database (MIT-BIH) 資料庫所提供的心電訊號進行最佳化的過程。在演算法效能驗證時，也會用PhysioNet提供的 Apnea-ECG database中長達124小時的心電訊號作評估。最後也會和其他R波偵測演算法做比較，探討效能的差異及造成的原因。

Heart disease is one of the leading causes of death, and the monitoring of the ECG signal is an essential prerequisite for judging whether there is a possibility of heart disease. The most apparent waveform in the ECG signal is the R wave, so the detection and interpretation of the R wave is the first step in analyzing the ECG signal. We proposed a time-domain multi-level R-peak detection algorithm for ECG signal processing. This method not only has low complexity but also maintains the shape of the waveform. The preservation of the ECG waveform is crucial to the following disease detection. This research optimizes the coefficients of the proposed algorithm. The R-peak detection algorithm is essentially a binary classifier. Therefore, in the process of optimizing the coefficients, the graphical method of classifier performance evaluation is used as the primary tool. This research uses the ECG signal from the MIT-BIH arrhythmia database (MIT-BIH) provided by PhysioNet for the optimization process. In the algorithm performance verification, 124 hours of ECG signals from the Apnea-ECG database provided by PhysioNet will also be used as well. Finally, the proposed algorithm will be compared with other R-wave detection algorithms to discuss the difference in performance and the reasons behind it.

[1] Department of Statistics of Ministry of Health and Welfare, “108年國人死因統計結果,” https://www.mohw.gov.tw/cp-16-54482-1.html (accessed Aug. 5th, 2020)
[2] G. F. Nemet, A. J. Bailey, “Distance and health care utilization among the rural elderly,” in Social Science & Medicine, May 2000, vol. 50, issue 9, pp. 1197-1208.
[3] Y. R. Tsai, Z. Y. Chang, C. W. Huang, “Time-domain multi-level R-peak detection algorithm for ECG signal processing,” in Biomedical Engineering, Healthcare and Sustainability (ECBIOS), 2019, pp. 35-38.
[4] J. Malmivuo and R. Plonsey, Bioelectromagnetism: Principles and Applications of Bioelectric and Biomagnetic Fields, Oxford University Press, New York, United States of America, 1995.
[5] Wikipedia contributors, “Electrocardiography,” Wikipedia, last modified date Jul. 25th, 2020, https://en.wikipedia.org/wiki/Electrocardiography (accessed Aug. 5th, 2020)
[6] C. W. Huang, “Implementation of an arrhythmia detection system for Android-based smartphones by using the morphological features of ECG signals,” M.S. thesis, National Tsing Hua University, Hsinchu, Taiwan, 2019.
[7] M. Cadogan, “ECG Lead Positioning,” Life in the Fastlane, last modified date Aug. 15th, 2019, https://litfl.com/ecg-lead-positioning/ (accessed Aug. 5th, 2020)
[8] M. Tanveer and Ram Bilas Pachori, Machine Intelligence and Signal Analysis, Springer, 2018.
[9] Ed Burns, “Right Bundle Branch Block (RBBB),” Life in the Fastlane, last modified date Mar. 16th, 2019, https://litfl.com/right-bundle-branch-block-rbbb-ecg-library/ (accessed Aug. 5th, 2020)
[10] T. Saito and M. Rehmsmeier, “The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets,” in PLoS ONE 10(3): e0118432. 2015, https://doi.org/10.1371/journal.pone.0118432
[11] A. L. Goldberger, L. A. N. Amaral, L. Glass, J. M. Hausdorff, P. Ch. Ivanov, R. G. Mark, J. E. Mietus, G. B. Moody, C.-K. Peng, H. E. Stanley, “PhysioBank, Physio Toolkit, and PhysioNet: Components of a new research resource for complex physiologic signals,” Circulation, Jun. 2000, vol. 101, no. 23, pp. e215–e220.
[12] G. B. Moody, R. G. Mark, “The impact of the MIT-BIH arrhythmia database,” in IEEE Eng in Med and Biol 20(3):45-50, May-June 2001. (PMID: 11446209)
[13] J. Pan and W. J. Tompkins, “A real-time QRS detection algorithm,” in IEEE Trans. Biomed. Eng., Mar. 1985, vol. BME-32, no. 3, pp. 230–236.
[14] W. Jenkal, R. Latif, A. Toumanari, A. Dliou, O. El B’Charri, F. Maoulainine, “QRS detection based on an advanced multilevel algorithm,” in International Journal of Advanced Computer Science and Applications, 2016, vol. 7, no. 1.
[15] W. Jenkal, R. Latif, A. Toumanari, A. Dliou, O. El B’Charri, F. Maoulainine, “An efficient algorithm of ECG signal denoising using the adaptive dual threshold filter and the discrete wavelet transform,” in Biocybernetics and Biomedical Engineering, 2016, vol. 36, issue 3, pp. 499-508.
[16] W. Jenkal, R. Latif, A. Toumanari, A. Elouardi, A. Hatim, O. El B’Charri, F. Maoulainine, “Real-time hardware architecture of the adaptive dual threshold filter based ECG signal denoising,” in Journal of Theoretical and Applied Information Technology, 2018, vol. 96, no. 14.
[17] D. Sadhukhan, M. Mitra, “R-peak detection algorithm for ECG using double difference and RR interval processing,” in Procedia Technology, 2012, vol. 4, pp. 873-877.
[18] T. Penzel, G. B. Moody, R. G. Mark, A. L. Goldberger, J. H. Peter, “The apnea-ECG database,” Computers in Cardiology, 2000; 27:255-258.
[19] H. Sedghamiz, “Complete Pan Tompkins implementation ECG QRS detector,” MATLAB Central File Exchange, retrieved Aug. 7th, 2020, https://www.mathworks.com/matlabcentral/fileexchange/45840-complete-pan-tompkins-implementation-ecg-qrs-detector/