心律不整分類以及偵測一直以來都是心電圖處理中的主要議題。在此研究中，我們納入了五種心律不整用於測試分類系統。首先，本文提出一個修改過的動態心電圖模型，將原先的高斯函數取代為柯西函數，原因是柯西函數相較之下擁有更接近的波峰。接著將動態模型與非線性迴歸演算法結合，這個系統將嘗試逼近真實的心電圖訊號直到誤差達到最小值。這麼做的原因有二：第一，產生一組合成且乾淨的心電圖訊號，仍然能保有原始的特性。第二，結果所產生出的係數能夠在未來重複使用，作為簡單且容易取得的資料。在這之後，系統會從輸入中擷取特徵，特徵擷取的方法首先會將資料轉換至相位空間，在此訊號會變成一系列的軌跡，藉由計算經過的次數便能夠產生二維直方圖。由於某些帶有心律不整的心電圖在時域下看起來混亂且難以解讀，相位空間下的統計資料將會揭露出隱含的資訊。更進一步地說，因為軌跡重疊的原因，傳統二值化的方法可能會遺失穩態出現的頻率，並且忽略掉暫態的關鍵特徵。最後，基於心律不整的種類，系統會選擇偵測異位的心跳或是使用基於倒傳遞類神經網路的分類器分類異常心律。

Cardiac arrhythmia classification and detection algorithms have been a major topic for electrocardiography. In this research, we include five arrhythmias for the development of the classification system. Firstly, a modified dynamical electrocardiogram (ECG) model is proposed, which replaces the original Gaussian function with Cauchy function for a better approximation at the peak. After that, the model is deployed in a nonlinear function for fitting a real ECG signal with a minimal error. There are two purposes for this approximation: (1) to generate a synthetic, clean ECG with the original properties, (2) reusable coefficients are an easy access for the signals needed in future simulations. The feature extraction is achieved by converting input to coordinated bins, which then accumulate the numbers of visiting trajectories into a 2D histogram. Since some ECGs with arrhythmias are chaotic and difficult to interpret in the time domain, demonstrating biomedical data in the phase space with statistical information would further reveal some underlying characteristics. More specifically, the conventional binary approach might lose the visiting frequency of steady-state dynamics by missing the overlapping trajectories, and probably ignoring the critical feature without considering the transient one. Finally, based on the type of arrhythmia, a classifier using backpropagation neural network is chosen to detect ectopic beats or classify the arrhythmias.

[1] J. Pan and W. J. Tompkins, “A real-time QRS detection algorithm,” IEEE Trans. Biomed. Eng., vol. BME-32, pp.230-236, Mar. 1985.
[2] D. T. Kaplan, “Simultaneous QRS detection and feature extraction using simple matched filter basis functions,” Computers in Cardiology, pp. 503-506, Sep. 1990.
[3] P. Davey, “A new physiological method for heart rate correction of the QT interval,” Heart, vol. 82, pp. 183-186, Aug. 1999.
[4] G. B. Moody, R. G. Mark, A. Zoccola, and S. Mantero, “Derivation of respiratory signals from multi-lead ECGs,” Computers in Cardiology, vol. 12, pp. 113-116, 1985.
[5] G. B. Moody, R. G. Mark, M. A. Bump, J. S. Weinstein, A. D. Berman, J. E. Mietus, and A. L. Goldberger, “Clinical validation of the ECG-derived respiration (EDR) technique,” Computers in Cardiology, vol.13, pp. 507-510, 1986.
[6] P. E. McSharry, G. D. Clifford, L. Tarassenko, and L. A. Smith, “A dynamical model for generating synthetic electrocardiogram signals,” IEEE Trans. Biomed. Eng., vol. 50, no. 3, pp. 289-294, Mar. 2003.
[7] G. D. Clifford and P. E. McSharry, “Generating 24-hour ECG, BP and respiratory signals with realistic linear and nonlinear clinical characteristics using a nonlinear model,” Computers in Cardiology, pp. 709-712, Sep. 2004.
[8] M. S. Billah, T. B. Mahmud, F. S. Snigdha, and M. A. Arafat, “A novel method to model ECG beats using Gaussian functions,” BMEI, pp. 612-616, Oct. 2011.
[9] S. Parvaneh and M. Pashna, “Electrocardiogram synthesis using a Gaussian combination model (GCM),” Computers in Cardiology, pp. 621-624, 2007.
[10] Y. Lu and Y. Yam, “A generalized ECG dynamical model with asymmetric Gaussians and its application in model-based ECG denoising,” BMEI, pp. 1-5, Oct. 2009.
[11] A. Orozco-Duque, F. J. Martinez-Tabares, J. Gallego, C. A. Rodriguez, I. D. Mora, G. Castellanos-Dominguez, and J. Bustammante, “Classification of premature ventricular contraction based on discrete wavelet transform for real time applications,” PAHCE, pp. 1-5, May 2013.
[12] F. Melgani and Y. Bazi, “Detecting premature ventricular contractions in ECG signals with Gaussian process,” Computers in Cardiology, pp. 237-240, Sep. 2008.
[13] J. J. A. Nasiri, M. Naghibzadeh, H. S. Yazdi, and B. Naghibzadeh, “ECG arrhythmia classification with support vector machines and genetic algorithm,” Computer Modeling and Simulation, pp. 187-192, Nov. 2009.
[14] K. Polat and S. Gunes, “Detection of ECG arrhythmia using a differential expert system approach based on principal component analysis and least square support vector machine,” Applied Mathematics and Computation, vol. 186, no. 1, pp. 898-906, Mar. 2007.
[15] D. A. Coast, R. M. Stern, G. G. Cano, and S. A. Briller, “An approach to cardiac arrhythmia analysis using hidden Markov models,” IEEE. Trans. Biomed. Eng., vol. 37, no. 9, pp. 826-836, Sep. 1990.
[16] A. Dallali, A. Kachouri, and M. Samet, “Classification of cardiac arrhythmia using WT, HRV, and Fuzzy C-Means clustering,” Signal Processing: An International Journal, vol. 5, no. 3, pp. 101-109, 2011.
[17] S. M. Jadhav, S. L. Nalbalwar, and A. Ghatol, “Artificial neural network based cardiac arrhythmia classification using ECG signal data,” ICEIE, vol.1, pp. 228-231, Aug. 2010.
[18] G. K. Prasad and J. S. Sahambi, “Classification of ECG arrhythmias using multi-resolution analysis and neural networks,” Conference on Convergent Technologies for the Asia-Pacific Region, vol.1, pp. 227-231, Oct. 2003.
[19] A. Gacek and W. Pedrycz, ECG signal processing, classification and interpretation, London : Springer-Verlag London Limited, 2012.
[20] N. Goldschlager and M. J. Goldman, Principles of clinical electrocardiography, 13th ed., New Jersey: Prentice-Hall, 1989.
[21] ECG interpretation, Lippincott Williams & Wilkins, Philadelphia, 2008.
[22] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by back-propagating errors,” Nature, vol. 323, pp. 533-536, Oct. 1986.
[23] R. Rojas, Neural Networks: A systematic introduction, Berlin, Springer-Verlag, 1996.
[24] A. Amann, R. Tratnig, K. Unterkofler, “Detecting ventricular fibrillation by time-delay methods,” IEEE Trans. Biomed. Eng., vol.54, no. 1, pp. 174-177, Jan. 2007.
[25] H. K. Khalil, Nonlinear Systems, 3rd ed., New York: Macmillan Pub. Co., 1992.
[26] M. Roopaei, R. Boostani, R. R. Sarvestani, M. A. Taghavi, and Z. Azimifar, “Chaotic based reconstructed phase space features for detecting ventricular fibrillation,” Biomedical Signal Processing and Control, vol. 5, no. 4, pp. 318-327, May 2010.
[27] 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, and H. E. Stanley, “Physiobank, physiotoolkit, and physionet: components of a new research resource for complex physiologic signals,” Circulation, vol. 101, no. 23, pp. e215–e220, 2000.