ECG Signal Processing And Automatic Diagnosis

In this thesis, the pretreatment, analysis of wave of electrocardiogram (ECG) signal are relatively researched, and the calculation of parameters and the automatic diagnosis are analyzed. Based on these results, we realize the software system of ECG Processing, which is integrated and has the real function.Firstly, the lead systems for ECG sampling are introduced, simply on the relation of every lead. Then we analyze the interferences of ECG, and presented the method of interferences rejection, such as power line interference, muscle electricity and baseline wander. At the same time, study and simulation of removing ECG noise based on wavelet are presented. By using the canceller, muscle electricity and power line interference are preferably eliminated from the ECG signals.Secondly, the detecting of P, QRS, T wave based on differential coefficient is presented, and this method makes the onset, peak and offset point preferable effect of orientation. Wavelet transform and moving window integration are used to detect the onset and offset of QRS wave of ECG. Local transfer is used to detect P, T wave, which is simpler to deal with vast data.Thirdly, based on the detecting of ECG wave, we discuss the calculation of the time, swing and cardio-axes parameters from 12 leads, which is used in the software system. The automatic diagnosis of normal and abnormal ECG is realized, and by inspecting two groups of data, it indicates that abnormal ECG can be picked up, which can be used in clinic.Finally, the structure of the software system is introduced, which consists of simultaneous 12 Lead ECG, High Frequency Electrocardiogram, Heart Rate Variability and Vector ECG, and so on. The system concourses multi-analysis methods, and combines automatic with manual diagnosis, which can afford comparatively integrated diagnosis report.