Analysis Of ECG Signal And Software Implement

Electrocardiogram (ECG) is a synthetic reflection of the heart electricity on body surface produced by the excited of pacemaker, atrium and ventricle. It is valuable for the ECG diagnoses. In recent years, the combination of the modern information technology and modern medical science makes automatic diagnosis and analysis of ECG becoming one of the most successful application in biomedical field. For this reason, to develop and research ECG diagnosis system is of great significance.Rsearch works on signal preprocess, ECG analysis methods and the development of automatic ECG analyzer have been done in this dissertation. The currently preprocess and analysis methods of ECG signal are first summarized. Then two methods which are better applied to ECG preprocess are analyzed in detail. The methods of extracting ECG characteristic information are principally researched. The algorithm based on wavelet to identify and locate the ECG characteristic waves is presented. The detect criteria of cha-racteristics wave is based on the relationship of singular point of signal and its wavelet transform. In order to identify and locate of low amplitude and multi-shape T wave, wave shape is determined firstly, then onset and offset is located secondly, and then different methods are applied in response to different shapes. The precision to locate T waves is greatly improved by this algorithm. On the basis of the above methods, the improved DTW (Dynamic Time Warping) is used to recognize ST shape. It overcomes the limitations that the traditional DTW algorithm uses the Euclidean distance as the distance measure, it uses the amplitude information, first order derivative (reflecting the change trend of wave), second derivative (reflecting the extremum information), makes the recognition more accuracy. Lots of experiments using Matlab and ECG data from MIT-BIH database have done to verify the accuracy and feasibility of the proposed algorithms.The ECG automatic analyzing software is developed with VC. It consists case history database manager, ECG wave display, ECG characteristic detecting, wave classify, case report printing and so on.