ANN BASED HYBRID APPROACH FOR ECG ANALYSIS

Abstract

Electrocardiogram (ECG) is an important bio-electrical signal which is extensively used by the cardiologists for diagnosing the state of cardiac system. An ECG signal recording can be decomposed into ECG cycles, and every ECG cycle can further be decomposed into simpler sub-patterns like P, Q, R, S, T peaks and PR and ST interwave segments. These peaks and interwave segments form the pattern vector of the ECG. So far, although a good amount of work has been done in this field, but still, no method has reached the stage of perfection. Therefore, there is utmost necessity to develop new methods which can be acceptable to experts (cardiologists) for computer aided interpretation of ECG signals. As a further step in this direction, the present work has been carried out for the development of neural network based hybrid approach for ECG analysis. This thesis meets the objectives of retrieving the R-peak of ECG signal for diagnostic purpose. ANN based hybrid approach has been used for ECG analysis. In this approach computational burden is reduced by using an algorithm based on linear approximation distance thresholding compression technique. The technique of artificial neural network based on Error Back Propagation (EBP) algorithm has been used to identify the R-peak from the compressed ECG data. The ANN topology 8-5-1' is used to detect the R-peak. This model employs a low scale of ANN. ANN based hybrid approach is implemented in Visual C++ 6.0 programming language using MFC (Microsoft foundation Classes) on Windows operating system