A Research Of ECG Detection And Classification Technology

Cardiac diseases are one of major diseases that endanger human's health. Analysis of ECG(Electrocardiogram) is an important means of clinical diagnosis of cardiac diseases. And it has been a research hotspot for many years at home and abroad that use computer to analyze ECG and diagnose cardiac diseases.QRS complex detection is the base of automatic analysis of ECG, which is mostly the first step of analysis and affects the following steps greatly. Multi-resolution processing can analyze signal at different scales and the amplitudes of different waveforms differ from each other at each scale for they have different frequency band. In the paper we process ECG signal using Marr wavelet and Mallat algorithm, and deduce a filter that can enhance QRS complex. Using this filter, peak detection and some other logic, an algorithm is developed that is able to locate the R-peak of QRS complex. High accuracy and medium computation load are the major features of our algorithm, which is certified by MIT-BIH arrhythmia database.Identification of abnormal ECG waveforms is a difficult problem of automatic analysis of ECG. Based on fuzzy set theory, we bring forward a new concept: similarity degree of waveforms, and use fuzzy pattern identification method to classify ECG waveforms. Different from traditional methods that view ECG as one-dimensional signal, ECG is treated as two-dimensional signal, an image, in this paper. Morphology has been used to process image for many years and the facts show that it is very effective. In my paper I give out a formula of calculating the morphological distance, Hausdorff distance, of two binary images, and then bring forward a definition of similarity degree using weighted H-distance. Also I use MIT-BIH database certify the definition and classification algorithm. The result shows that my definition of similarity degree is good and the algorithm is feasible.