Research On ECG Signal Processing Method Of Automatic External Defibrillator

Ventricular tachycardia (VT) and ventricular fibrillation (VF) are the most malignant arrhythmia, which are prone to cause a sudden death. It means the end of life if no treatment such as defibrillation is applied. So, it is important to develop a portable automatic external defibrillator (AED), which has the electrocardiogram (ECG) monitoring function. The AED can analyze ECG signal synchronously and capture the malignant arrhythmia promptly. If VF rhythm is appeared, the AED can provide the non-professionals the operation processes of defibrillation through intelligent voice prompt. A timely defibrillation can increase patients survival rate greatly.This paper.is centered on the AED. At first, analyze the characteristics of the ECG, the source of noises in the collected signals, as well as the design methods of digital filter, which is used to remove the baseline drift and power-line interference. Secondly, this paper describes the principle and shortcomings of the traditional differential threshold method to detect QRS complex in detail, and based on which, put forward an improved differential threshold method, then, compare the two methods by simulation results. Finally, this paper develops a more in-depth study on VF detection algorithm, which is the pivot of this paper. After analyzing and conducting experiments on two classic non-linear VF detection algorithms, that is, the complexity measure algorithm and approximate entropy algorithm, the sample entropy (SampEn) is proposed as a descriptor of VF detection algorithm. Experiment is carried out by the Creighton University Ventricular Tachyarrhythmia database (CUDB), which is a widely recognized and well annotated database.The quality parameters show that sample entropy algorithm can distinguish VF from non VF effectively, which has a high specificity and accuracy. The SampEn algorithm has a broad application prospects in the field of VF detection.