Automatical ECG Detection Based On Embedded System

The automated detection of electrocardiogram (ECG) is to extract the characteristic features of ECG by computer or electrocardiograph automatically. As a faint bioelectric signal, the ECG signal exhibit obvious unsteadiness and contain many noises. So an algorithm of detecting ECG characteristic based on wavelet transform, which is also combined with simply classing for ECG signals, is implemented. The algorithms that could successfully remove various noises are introduced to decompose the ECG signals on multi-scales. In characteristic scales, the electrocardiac signals can be decomposed. Though the ECG feature extraction was studied by many researchers, the method of automated detection of ECG is still need to be improved. Based on the demands of academic research and application, this dissertation carries on the algorithms study on improving the precision of the delineation algorithms. The characteristic features of ECG, including the peaks and limits of the individual QRS waves, P and T waves and the intervals of ST-segment, are further researched and experimented. This paper mainly includes the following distinctive works accomplished:1. A 12-lead synchronous ECG database was constructed in VC++ platform, which made it possible to use 12-leads ECG signals with higher frequency to estimate this algorithm, and had been prepared for ECG automated diagnosis.2. In this thesis, a robust single–lead ECG delineation system based on the wavelet transform was developed. And an algorithm to extract the features of ECG was simulated in Matlab platform which was used as calculation engine also. The algorithm was evaluated on several manually annotated databases, such as MIT-BIH Arrhythmia, and European ST-T databases. The algorithms also fit arbitrary lead's signal from 12-leads synchronous electrocardiograph. The QRS detector obtained a precision above 99.8%. So all conclusion prove that this algorithm is reliable. 3. To enhance detection accuracy of P waves in pathologic ECG waves, a method of ECG classified combined with the delineation of P wave was put forward.4. It is approved that this algorithm is transplantable while it has been embedded into 12-leads synchronous electrocardiograph and achieves reliable results.