臺灣博碩士論文加值系統

本論文提出利用自動迴歸移動平均具外部輸入的數學模型來模擬正常的心電圖，其中外部輸入是由三個主要元件組成，此三個主要元件分別為右三尖瓣膜、心室間隔以及主動脈瓣膜。估測數學模型的係數將採用最小平方誤差及基因法則，最小平方誤差所估測的參數為區域最佳解，而基因法則所估測的參數較接近全域最佳解，是兩者間最主要的差異。但是利用最小平方誤差產生的數學模型的係數，可將心電圖其中重要的J點突顯出。本論文亦呈現不同變化的外部輸入組合對心電圖的影響效果，希望這些影響效果及根據數學模型所估測的係數可以提供新的方向，在未來能夠預測可能的心臟缺陷。

The auto-regressive exogenous input moving average (ARMAX) model is proposed to approximate and to predict the normal electrocardiogram (ECG) in the thesis. The most important X part of the ARMAX is consisted of three major components, such as the right tricuspid valve, the interventricular septum, and the aortic valve. Both minimum square error (MSE) criterion and Generic Algorithm (GA) have been implemented to estimate the coefficients of the ARMAX. The major difference is that the predicted results of MSE criterion usually are local optimal, and the results of GA is superior to that of MSE criterion. Nevertheless, the J junction can be clearly observed from the results of the MSE criterion. Effects by varying the X part are also demonstrated. We hope that the proposed ARMAX model can offer a new approach to diagnose the potential heart defects in the future.

ABSTRACT (IN CHINESE)
ABSTRACT (IN ENGLISH)
TABLE OF CONTENTS
LIST OF FIGURES
LIST OF TABLES
CHAPTER 1 Introduction
CHAPTER 2 Electrocardiogram Waves
CHAPTER 3
3.1 Hamming window
3.2 ARMAX
CHAPTER 4 Results and Discussions
4.1 ARMAX with Original Data
4.2 ARMAX with New Locations and Original Data
4.3 ARMAX with Smoothing Data
CHAPTER 5 Conclusion Reference
Appendix

[1] M.D. Aswini Kumar. “ECG - simplify.’’ . 2008.
[2] E. Braunwald, “Heart Disease: A Textbook of Cardiovascular Medicine, Fifth Edition.” pp. 108, 1997.
[3] C Van Mieghem, M Sabbe, D Knockaert, “The clinical value of the ECG in noncardiac conditions.” pp. 1561-76. 2004.
[4] JH Chesebro, U Rauch, V Fuster, JJ Badimon, “Stabilization of the Patient With Acute Coronary Syndromes.” American Heart Association. Nov 28, 2005.
[5] P Chazal, M O’Dwyer, RB Reilly. “Automatic classification of heartbeats using ECG morphology and heartbeat interval features.”pp.1196-1206. IEEE 2004
[6] I Christov, G Gómez-Herrero, V Krasteva, I Jekova, A Gotchev, and K Egiazarian, “Comparative study of morphological and time-frequency ECG descriptors for heartbeat classification,”. pp. 876-887. 2006.
[7] G Moody, R Mark. “QRS morphology representation and noise estimation using the Karhunen-Lo`eve transform.” pp.269-272. 1989.
[8] P Laguna, R Jane, S Olmos, NV Thakor, H Rix, P Caminal. “Adaptive estimation of QRScomplex wave features of ECG signal by the Hermite model.” pp.58–68. 1996.
[9] M Lagerholm, G Peterson, G Braccini, L Edenbrandt, L S¨ornmo. “Clustering ECG complex using Hermite Functions and self-organizing maps.” IEEE pp. 838-848.2000.
[10] C.W. Li, C.X. Zheng, and C.F. Tai, “Detection of ECG characteristic points using wavelet transform,” IEEE vol. 42, pp. 21-28. 1995.
[11] H.G. Hosseini, D. Luo, and K.J. Reynolds, “The comparison of different feed forward neural network architectures for ECG signal diagnosis.” pp. 372-378, vol. 28, 2006.
[12] M. Shahram and K. Nayebi, “ECG beat classification based on a Cross-Distance analysis,” pp. 234-237. 2001.
[13] R. D. Throne, J. M. Jenkins, and L. A. Dicarlo, “A comparison of four new time domain techniques for discriminating monomorphic ventricular tachycardia from sinus rhythm using ventricular waveform morphology” IEEE. pp. 561-570.vol. 38, 1991.
[14] D. Lin. L. A. DiCarlo. and J. M. Jenkins, "Identification of vcntric-I201 G F. Tomaselli. A. P. Nielscn. W. L. Finke. L. Sengupta. J. C. Clark. and J. C. Griffin. "Morphologic diffkences of the endocardia1 electmgram in beats of sinus and ventricularorigin." PACE. vol. I I. pp. 254-262. Mar. 1988.
[15] L. Ljung, “System identification, 2nd edition.” PTR: Prentice-Hall, 1999.
[16] Chin-Teng Lin and C.S. George Lee, “Neural fuzzy systems”, 2003.