Research On Real Time Automatic Analysis Of ECG Signals Based On The Community-Monitor Mode

The community medicine has attracted very much attention as a new medicine mode in the world and has very good foreground . This new medicine mode has brought forward new demand to the automatic analysis of ECG signals. In this study, I have used the way of wavelet transform (WT) to study the real-time automatic analysis of ECG signals.In the past several ten years, many people has made great efforts to detect and analysis the ECG waves with vary methods. But none of those methods could satisfy the clinical demand, and the most of those algorithms are after-analysis algorithm based on PC, it cann't satisfy the demand of community-monitor. In the same time, the community-monitor requires the automatic analysis technology much more because of the particularity that community-monitor differs from the in-hospital-monitor. It is very important for the patient who has disease that can endanger the patient's life.The existent algorithms often put the importance on the detection of QRS complex, there is few researcher to study the detection of ST-T segment. The analysis of ST-T segment is more important actually to community-monitor. In this study, I adopted the way of wavelet transform to process ECG signals. First of all, the algorithm based on WT eliminated the three primary noise (including power line interference, electrical interference of muscle and baseline wander), and then detected the character of QRS complex. In the end, I study the way of ST-T segment detection based on the QRS character and brought forward a algorithm for the ST-T detection.In this study, a lot of experiments with the data from MIT-BIH database and ECG simulator validated the accuracy and feasibility of the proposed algorithm.The detection algorithm of ST-T segment is a novel method ,there is no same method reported in the world up to now.