The Research Of ECG Signal Automatic Diagnosis System

With the development of society and the continuous uplifting of people’s living standards in recent years, there is growing awareness of human beings health. Heart disease is a common disease and one of the major diseases that endanger human’s health. The electrocardiogram automatic diagnosis as people’s research focus, it will be a great fruit in medical field if it can be put into practice. The ECG automatic diagnosis system of this research can detect each ECG waveform correctly, calculate ECG waveform characteristic parameter values, use the given automatic diagnostic criterion to diagnosis one ECG is normal or not automatically.In the thesis, firstly, the generating principle and characteristics of ECG are introduced, and then we summary the standard12-lead system which commonly used in clinical. Then we analyze the interference and noise, and presented the method of interferences rejection for baseline drift, denoised signal can effectively filter out interference as well as remain useful information; against the requirements for real-time and accuracy of the detection algorithm, the zero crossing detection technology are improved for multiple detection events, then used to detect location of R wave. Our experiments show that the algorithm is very fast and the detection accuracy is high; method based on the slope is used to detect Q、S wave; method based on the amplitude is used to detect P、T wave; then according to the experience of the expert, we select12ECG characteristic parameters as the standard of automatic diagnosis to distinguish normal and abnormal ECG.Finally, the structure of the automatic diagnosis system is introduced, the software can conduct a series of operations on input ECG data automatically, such as preprocessing, feature extraction, parameter calculation, and eventually arrive at a diagnosis. Through two sets of actual ECG test, test results show that the system can well distinguish between normal and abnormal ECG.