臺灣博碩士論文加值系統

心電圖為記錄人體心肌細胞電性改變後，身體體液電位變化以非侵入方式擷取之資訊。數醫學ECG信號數據庫中顯示，每個人的ECG信號均極複雜且隨機變化，由於ECG信號的這種特質，因此個人ECG信號很難以人為方式變造複製。最近幾年國&;#20869;外研究人員對生命循環系&;#32479;的核心-ECG的非線性特性作了深入研究，研究論文顯示，心電訊號看似複雜，但卻顯現混沌動態特質，也就是ECG雖為非線性動態行為但具有特殊規律性。本研究嘗試以非線性心電訊號為研究對象，規劃此一研究主題，提出一種使用混沌心電訊號進行個人身份辨識的系統。利用心電訊號在時域上的電壓有效值，並且將心電訊號轉至相平面並利用混沌理論分析心電訊號狀態的關聯維數與李亞普若夫(Lyapunov)指數譜等三種特徵值，結合神經網路學習與辨識功能在人們相同生理條件下，所反映的心臟動力學整體特性，作為個人身份生物辨識的方法。希望本研究所提出的方法，未來可作為一種個人身分辨識的工具。

Electrocardiography (ECG) is a noninvasive recording featured by electrocardiographic devices which is a transthoracic interpretation of the electrical activity of the human’s heart over time captured. From the data base of medical ECG signals for people, ECG seems to be irregular, random, and changing from person to person. Because of high randomness and complexity of ECG in human beings, its feature is extremely hard and is likely impossible to be duplicated artificially. However, it has recently been shown in the literature that the kind of signals is, in fact, chaotic. Because people’s ECGs are extremely hard to be artificially duplicated, thus intends to investigate the way to extract the specific biometric features of ECG signals for possible use in the biometric personnel identification. The root mean square of the ECG in the time domain is viewed as one of the characteristic values. The signal is converted into the phase plane and ECG chaos extractor is applied to capture the major indices of ECG chaos, i.e. Lyapunov exponents spectrum and correlation dimension. The above mentioned root mean square, Lyapunov exponent and correlation dimension are used as the key input variables for neural networking training and used in the identification scheme.

誌謝辭 i
摘要 ii
Abstract iii
Contents iv
List of Figures vii
List of Tables x
Chapter 1 Introduction 1
Chapter 2 ECG and System Description 4
2.1 Introduction 4
2.2 Electricity Transmit in the Cells 4
2.3 Each Waves 5
2.3.1 P Wave 5
2.3.2 QRS Wave 5
2.3.3 T Wave 6
2.3.4 Each Interval 6
2.4 Leads 7
2.4.1 Standard-lead 7
2.4.2 Augmented-lead 8
2.4.3 Chest-lead 8
2.5 Application of the ECG 8
2.6 System Description 9
2.6.1 Device 9
2.6.2 ET-600 10
2.6.3 NI USB-6211 10
2.6.4 Digital Filter 10
Chapter 3 Chaotic Analysis for ECGs 11
3.1 Time Series 11
3.2 Attractor 11
3.3 Time-Delay Space Reconstruction 11
3.3.1 Time-Delay 12
3.3.2 Embedded Dimension 12
3.4 Correlation Dimension 13
3.5 Lyapunov Exponent 14
3.5.1 Definition of the Lyapunov Exponent 14
3.5.2 Calculating the Lyapunov Exponent 15
Chapter 4 Preprocessing and Classification 17
4.1 Signal Pre-Processing 17
4.1.1 Outliers 17
4.1.2 Standard Deviation 17
4.2 Neural Network 18
4.2.1 Principle of the Neural Network 18
4.2.2 Back Propagation Neural Network 18
4.2.3 The Algorithm Resilient Backpropagation 20
4.2.4 Processing of the Back Propagation Neural Network 21
Chapter 5 Experimental Results 23
5.1 How to Obtain the ECG 23
5.2 Analyzing the ECG 24
5.3 Deleting the Pattern of the Outliers 26
5.4 Training and Testing for Neural Network 26
Chapter 6 Conclusion and Future Work 28
6.1 Conclusion 28
6.2 Future Work 29
References 30

[1]K. T. Lai and K. L. Chant, “Real-time classification of electrocardiogram base on fractal and correlation analyses,” Proceedings of the IEEE Engineering in Medicine and Biology Society, pp. 119-122, 1998.
[2]M. G Tsipouras, D.I. Fotiadis and D. Sideris, “Arrhythmia classification using the RR-interval duration signal,” Computers in Cardiology, pp. 485-488, 2002.
[3]P. de Chazal, B. G. Celler and R. B. Reilly, “Using wavelet coefficients for classification of the electrocardiogram,” Proceedings of the IEEE Engineering in Medicine and Biology Society, pp. 64-67, 2000.
[4]Y. Gang and M. Malik, “Heart rate variability analysis in general medicine,” Indian Pacing and Electrophysiology Journal, vol. 3, pp. 34-40, 2003.
[5]Y. Gu, J. L. Duan and Y. C. Wang, “Heart rate variability and the progress of its clinical application,” International Journal of Cardiovascular Medicine, vol. 9, pp.17-19, 2008.
[6]A. Babloyantz and A. Destexhe, “Is the normal heart a periodic oscillator?” Biological Cybernetics, vol. 58, pp. 203-211, 1988.
[7]O. Fojt and J. Holick “Applying nonlinear dynamic to ECG signal processing,” IEEE Engineering in Medicine and Biology Magazine, vol. 17, pp. 96-101, 1998.
[8]S. Lee, T. L. Kang, H. Y. Quan and X. Tian, “Analysis of HRV based on correlation dimension and largest Lyapunov exponent,” Journal of Biomedical Engineering Research, vol. 28, pp. 188-192, 2009.
[9]P. B. Pascolo, A. Marini, R. Carniel and F. Barazza, “Posture as a chaotic system and an application to the Parkinson’s disease,” Chaos, Solitons & Fractals, vol. 24, pp. 1343-1346, 2005.
[10]B. Anuradha and V. C. Veera Reddy, “ANN for classification of cardiac arrhythmia,” ARPN Journal of Engineering and Applied Sciences, vol. 3, pp. 1-6, 2008.
[11]C. Raab, J. Kurths, A. Schirdewan and N. Wessel, “Normalized correlation dimension for heart rate variability analysis,” Biomedizinische Technik Technik, vol. 51, pp. 229-232, 2006.
[12]A. Kikuchi, T. Shimizu, A. Hayashi, T. Horikoshi, N. Unno, S. Kozuma and Y. Taketani, “Nonlinear analyses of heart rate variability in normal and growth-restricted fetuses,” Early Human Development, vol. 82, pp. 217-226, 2006.
[13]AS Al-Fahoum and AM Qasaimeh, “ECG arrhythmia classification using simple reconstructed phase space approach,” Computers in Cardiology, vol. 33, pp. 757-760, 2006.
[14]Q. Jiao, Y. X. Guo, W. F. Cao and Z. G. Zhang, “Short time ECG signal anaylsis based on the reconstruction of phase space,” Chinese Journal of Medical Instrumentation, vol. 32, pp. 257-264, 2008.
[15]L. Biel, O. Pattrsson, L. Philipson and P. Wide, “ECG analysis: a new approach in human identification,” IEEE Transaction on Instrumentation and Measurement, vol. 50, pp. 808-812, 2001.
[16]H. I. Eisen, “Surgical ventricular reconstruction for heart failure,” The New England Journal of Medicine, vol. 360, pp. 1781-1754, 2009.
[17]T. W. Shen, W. J. Tompkins and Y. H. Hu, “One-lead ECG for identity verification,” Proceedings of the Second Joint EMBS/BMES Conference, pp. 62-63, 2002.
[18]S. A. Israel, J. M. Irvine, A. Cheng, M. D. Wiederhold and B. K. Wiederhold, “ECG to identify individuals,” Pattern Recognition, vol. 38, pp. 133-142, 2005.
[19]F. Agrafioti and D. Hatzinakos, “ECG based recognition using second Order Statistics,” Proceedings of the Communication Networks and Services Research Conference, pp. 82-87, 2008.
[20]S. Z. Fatemian and D. Hatzinakos, “A new ECG feature extractor for biometrics recognition,” Proceedings of the 16th International Conference on Digital Signal Processing, pp. 1-6, 2009.
[21]D. Davis, How to Quickly and Accurately Master ECG Interpretation, Better World Books, Mishawaka, 1985.
[22]F. Takens, “Determining strange attractors in turbulence,” Lecture Notes in Math, vol. 898, pp. 245-262, 1989.
[23]M. Hénon, “A two-dimensional mapping with a strange attractor,” Communications in Mathematical Physics, vol. 50, pp. 69-77, 1976.
[24]P. Grassberger and I. Procaccia, “Measuring the strangeness of strange attractors,” Physica A, vol. 9, pp. 189-208, 1983.
[25]M. Sano and Y. Sawada, “Measurement of the Lyapunov Spectrum from a Chaotic Time Series,” Physical Review Letters, vol. 55, pp. 1082-1085, 1985.
[26]R. R. Hashemi, L, A, Blanc, C. T. Rucks and A. Rajaratnam, “A hybrid intelligent system for predicting bank holding structures,” European Journal of Operational Research, vol. 109, pp. 390-402, 1998.