臺灣博碩士論文加值系統

典型心房撲動、非典型心房撲動及心房顫動為不同生理機制之心房心律不整，在發生此三類心律異常時，心房之電活化現象變得不規則且停止同調性的收縮，然而，藉由體表心電圖之判讀我們不易將此三類作清楚的區分。非線性理論近年來常被應用於心臟訊號且心房之心律異常被證實具有非線性混沌系統之特徵，於此研究，我們建構出一套非線性分析法來分析典型、非典型心房撲動及心房顫動之體表心電訊號，找出訊號之非線性特性進而加以分類。方法: 本研究中，我們取得一筆正常心律、二十筆心房撲動(十筆典型、十筆非典型)及七筆心房顫動之十二導程心電圖，利用獨立成份分析法由十二導程心電訊號中萃取出心房之心電訊號，求得心房訊號之四個非線性特徵參數，相關維度、最大李雅普諾夫指數、柯爾莫哥羅夫熵及複雜度，最後利用此四個非線性參數建構出一個類神經網路來將此三類之心律異常疾病加以分類。結果: 在此三類心律異常疾病中，典型心房撲動呈現較小之非線性參數數值，其相關維度、最大李雅普諾夫指數、柯爾莫哥羅夫熵及複雜度範圍分別為1.76~4.36, 0.06~6.92,0.06~9.74,及84~115，而在心房顫動時，心電訊號變得較混亂，非線性參數呈現較大之數值，其相關維度、最大李雅普諾夫指數、柯爾莫哥羅夫熵及複雜度範圍分別為4.56~6.34，11.06~20.58，15.63~33.05，及124~158，在非典型心房撲動發生時，其心電訊號較心房顫動規則且較典型心房撲動複雜，由結果得知其非線性參數界於典型心房撲動及心房顫動之間，此外，我們利用統計分析法證實出此四個非線性特徵值於此三類之心律異常疾病中的確是具有分類上的差異，且在類神經網路分類上，也確實可以將此三類作良好的區分。結論: 藉由非線性分析法，我們觀察到典型心房撲動、非典型心房撲動及心房顫動體表心電訊號之非線性參數具有不同的特性，依此特性我們可將此三類心律異常疾病作清楚的分類。

Atrial flutter (AFL, including typical and atypical) and atrial fibrillation (Af) have different generating mechanisms in atrium. However, they are often cross-classified on the surface ECG. During AFL and Af, atrial activities appear disordered and lose the regular rhythm. Nonlinear analysis has recently been applied to electrograms, and atrial arrhythmia has shown evidence that indicates the possibility of deterministic chaos. In this study, we applied methods from the theory of nonlinear dynamics to discriminate electrograms of atrial flutter and fibrillation in humans. Methods: Atrial activities were obtained by using independent component analysis of standard 12-lead surface electrocardiograms. One sinus rhythm, twenty AFLs (10 typical and 10 atypical) and 7 Afs were analyzed. The Grassberger-Procaccia algorithm was applied to estimate the correlation dimension (D2). Three other nonlinear parameters, largest Lyapunov exponent (□1), Kolmogorov entropy (KE), and Lempel-Ziv complexity (C) were also calculated. Finally, the four nonlinear characteristics were considered to classify these three types of arrhythmias by using the neural network. Results: In these three types of arrhythmias, for typical flutter, nonlinear parameters were relatively smaller. D2, □1, KE, and C ranged from 1.76 to 4.36, 0.06 to 6.92, 0.06 to 9.74, and 84 to 115, respectively. For Af, ECG patterns become chaotic and these parameters presented higher values. D2, □1, KE, and C ranged from 4.56 to 6.34, 11.06 to 20.58, 15.63 to 33.05, and 124 to 158, respectively. During atypical flutter, ECG shows more complex than typical flutter but more regular than Af and thus the magnitude of these nonlinear parameters presented between typical flutter and Af. Statistical analysis showed evidence that these parameters exhibited a significant differentiation allowing the classification of these arrhythmias. By using the neural network classification, we also obtained a desirable result. Conclusion: The nonlinear analysis provided us an advantageous technique to discriminate among typical flutter, atypical flutter and Af electrograms from surface ECG.

[1] A. Babloyantz, A. Destexhe, “Is the normal heart a periodic oscillator?,” Biol. Cybern., 58: 203-211, 1998.
[2] B. P. Hoekstra, C. G. Diks, M. A. Allessie, J. DeGoede, “Nonlinear analysis of the pharmacological conversion of sustained atrial fibrillation in conscious goats by the class Ic drug cibenzoline,” Chaos, 7: 430-446, 1997.
[3] A. Casaleggio, M. Morando, S. Pestelli, S. Ridella, “Study of the correlation dimension of ECG signals based on MIT-BIH arrhythmia data base ECGs,” IEEE Comp. Soc., 401-404, 1990.
[4] A. Casaleggio, M. Rabbia, C. Lamberti, G. Bortolan, ”Study of the influence of waveform variations on the ECG correlation dimension,” IEEE Comp. Soc., 601-604, 1992.
[5] A. Casaleggio, “Differences on the correlation dimension of MIT-BIH ECG database recordings,” IEEE Comp. Soc., 539-542, 1993.
[6] G. Bortolan, A. Casaleggio, “The correlation dimension in rest ECG: a study in normal and arrhythmic patients,” IEEE Comp. Soc, 393-396, 1995
[7] A. Casaleggio, S. Braiotta, “Estimation of Lyapunov exponents of ECG time series: the influence of parameters,” Chaos, Solitions & Fractals, 8: 1591-1599, 1997.
[8] A. Casaleggio, S. Braiotta, A. Corana, “Study of the Lyapunov exponents of ECG signals from MIT-BIH database,” IEEE Comp. Soc., 697-700, 1995.
[9] M. I. Owis, A. H. Abou-Zied, A. M. Youssef, Y. M. Kadah, “Study of features based on nonlinear dynamical modeling in ECG arrhythmia detection and classification,” IEEE Trans. Biomed. Eng., 49: 733-736, 2002.
[10] X. S. Zhang, Y. S. Zhu, “Detecting ventricular tachycardia and fibrillation by complexity measure”, IEEE Trans. Biomed. Eng., 548-555, 1999.
[11] E. G. Daoud, F. M. Morady, “Pathophysiology of atrial flutter,” Annu. Rev. Med., 77-83, 1998.
[12] M. A. Allessie, W. J. E. P. Lammers, F. I. M. Bonke and J. Hollen, “Experimental evaluation of Moe’s multiple wave hypothesis of atrial fibrillation,” Cardiac Electrophysiology and Arrhythmias, 265-275, 1985.
[13] Jian-Hung Liu, Tsair Kao, “Application of independent component analysis to the characterization of atrial fibrillation,” Thesis, Institute of Biomedical Engineering National Yang-Ming University, 2003.
[14] B. P. T. Hoekstra, C. G.H.Diks, M. A. Allessie, and J. DeGoede, “Nonlinear analysis of epicardial atrial electrograms of electrically induced atrial fibrillation in man,” J. Cardiovasc. Electrophysiol., 6:419-440, 1995.
[15] P. Grassberger, T. Schreiber, C. Schaffrath, “Nonlinear time sequence analysis,” Int. J. Bif. Chaos, 1: 521-547, 1991.
[16] J. Theiler, “Spurious dimension from correlation dimension algorithms applied to limited time-series data,” Phys. Rev. A., 34: 2427-2432, 1986.
[17] M. Sano, Y. Sawda, “Measurement of the Lyapunov spectrum form a chaotic time series,” Phys. Rev. Lett., 55: 1082-1085, 1985.
[18] A. Lempel, J. Ziv, “On the complexity of finite sequence,” IEEE Trans. Inform., Theory, IT22: 75-81, 1976.
[19] R. Hegger, H. Kantz, T. Schreiber, “Nonlinear time series analysis,” TISEAN 2000, http://www.mpipks-dresden.mpg.de/~tisean\TISEAN_2.1\index.html
[20] J. J. Rieta, J. Millet, V. Zarzoso, F. Castells, C. S''anchez, R. Garcia, S. Morell, “Atrial fibrillation, atrial flutter and normal sinus rhythm discrimination by means of blind source separation and spectral parameters extraction”, IEEE Computers in Cardiology, 25-28, 2002.
[21] J. J. Rieta, F. Castells, C. S''anchez, J. Igual, “ICA applied to atrial fibrillation analysis,” 4th International Symposium on ICA and BSS.
[22] A. M. Fraser, H. L. Swinney, “Independent coordinates for strange attractors from mutual information,” Phys. Rev. A., 33: 1134-1140, 1985.
[23] M. B. Kennel, R. Brown, and H. I. Abarbanel, “Determining embedding dimension for phase-space reconstruction using a geometrical construction”, Phys. Rev. A., 45: 3403-3411, 1992.
[24] Ann-Shin Liu, Tsair Kao, “Analysis of atrial electrograms in human atrial fibrillation,” Thesis, Institute of Biomedical Engineering National Yang-Ming University, 1999.