| Electrocardiogram (ECG) workstation is an advanced ECG examination system which achieves digital medical treatment using several technologies, such as signal processing, modern transmission, network, database, display technology and so on. ECG workstation provides a simple, effective, exact treatment method for doctors. However, there is a relatively great gap between our nation and abroad in the research on ECG workstation. The purpose of this thesis is to come out an ECG automatic analysis software which has self-owned intellectual property rights. Based on VC++ 6.0, we realize ECG preprocessing, frequency domain ECG analysis, heart rate variability analysis and high frequency ECG analysis. Moreover, we research on electrocardiogram automatic analysis algorithm, and test the effect of our software and algorithm based on ECG database such as MIT-BIH and Fantasia database.In this thesis, we design a digital trap to eliminate power-line interference, reduce baseline wander and remove high frequency noise with wavelet transform. Meanwhile, we try to find the characteristic points of QRS complexes accurately.A software for heart rate variability analysis in time domain and frequency domain has been designed based on VC++ 6.0.We propose a new method to determine whether heart rate variability signal is chaos or not by means of Cao algorithm.Also, a stability criterion is proposed to determine minimum embedding dimension, the effect of this criterion has been tested.We advance a new HRV clinical index ,test its reliability and sensibility through Fantasia database. We also find and implement the best phase space reconstruction method for HRV, calculate Lyapunov index. Furthermore, we implement frequency domain ECG (FECG) analysis software based on VC++ 6.0, advance n segments method for FECG function, propose modified preprocessing method and modified coherence function, and also propose new indices for FECG Finally, we implement high frequency ECG analysis software based on VC++ 6.0. |
PDF Full Text Request