The System Of ECG Monitoring And Auxiliary Diagnosis Of Arrhythmia

Cardiac arrhythmia refers to any abnormal of the origin site, cardiac frequency and rhythm, and impulse conduction, often accompanied with a variety of organic heart disease, such as coronary heart disease, cardiomyopathy, myocarditis, and rheumatic heart disease. Each arrhythmia will cause changes in ECG It is an effective way to detect and diagnose of arrhythmia by using ECG At present, the diagnosis of arrhythmia mostly completed by clinician, It time consuming and depends on the doctor's personal business standards and accountability. It is very helpful to achieve automatic recognition of arrhythmias, on the one hand it provides emergency ambulance with heart disease and predicts the development of disease severity, obtains the best time of treatment; The other hand, it can greatly save doctors time and help doctors to take clinical diagnosis. Therefore, the automatic arrhythmia diagnosis has significant social and economic benefits. In this paper, the following work has been performed:1.Preprocessing of ECG signal. Using different methods to eliminate the interference of signal. The baseline drift is removed firstly by using wavelet decomposition and reconstruction method. then the wavelet adaptive threshold method is used to remove electromyography interference, and effectively overcome the Gibbs phenomenon. Finally, a 50HZ notch filter based on FIR is used to remove power-line interference.2.Eeature extraction. The filtered ECG signals are decomposed by aTrous algorithm using dyadic spline wavelets, and the signal's characteristic points detection theory of wavelet transformation (WT) is illustrated. Then the R wave peak point is extracted firstly, according to the location of the R wave peak point, the starting point and end point of QRS wave is extracted by using linear fitting method. Finally, according to set the appropriate search window, the P wave and T wave peak point, and then start point and end point is extracted by using linear fitting method.3.ECG monitoring. According to the existing ECG monitor and guidance of clinicians, the calculation method of ECG monitoring technical specifications is given.4.Automatic diagnosis of arrhythmia. According to ECG changes can differentiate and diagnose the type of. Arrhythmia. First of all, according to R wave peak and RR interval, the whole morphology of arrhythmic heartbeat is extracted from electrocardiograph dynamically, Then discrete cosine transform(DCT) and inverse discrete cosine transform are used to compress the data; Second, A BP neural network is designed for arrhythmic heartbeat recognition, and the initial weights and thresholds of network are optimized by DNA algorithm. Finally, the MIT/BIH ECG database is used to test the DNA-BP neural network. The experiment results demonstrate efficient of this approach.5.The system of ECG monitoring and auxiliary diagnosis of arrhythmia. The system consists of data management, ECG monitoring and arrhythmia diagnosis. The system interface is developed by using Visual C++, the system users and case databases is built by SQL SERVER 2000,and in the system the MATLAB engine is encapsulated,using MATLAB computing power to realize the diagnosis of arrhythmia.