Algorithms Study And Software Implentation For Automatic Detection Of ECG Signal

Heart disease is one of the most serious threats to human health, so its prevention and diagnosis has become a main problem of the medical profession. ECG (electrocardiogram, ECG) as an important basic for the diagnosis of cardiovascular diseases, the processing and analysis of ECG studies are becoming more and more important.Based on analysis of the ECG physiological causes, waveform and frequency domain characteristics, we make a simple and effective QRS identification method. Based on these, LVQ neural network has been used in the ECG waveform classification, it makes ECG signal automatic detection and analysis come true. Evaluate this algorithm with ECG signals of MIT-BIH standard database, the outcome indicate this method has well performance.Forthermore, the author has design software for ECG automatic processing and analysis which is helpful for clinical treatment. The software uses C + + language, and it can be achieved by C + + Builder compiler. It has implemented a lot of features, such as: the ECG signal preprocessing, the QRS automatically detect, waveform display, data storage and playback, and it can be applied to the general ECG monitoring and testing.In this paper, the innovative point is applied neural network LVQ to the ECG waveforms classification and has obtained good effects. As well, software has been made ultimately which can be used for a clinical ECG analysis and automatic detection.