Artificial neural network for ECG arryhthmia monitoring

Abstract

The application of a multilayer perceptron artificial neural network model (ANN) to detect the QRS complex in ECG (electrocardiography) signal processing is presented. The objective is to improve the heart beat detection rate in the presence of severe background noise. An adaptively tuned multilayer perceptron structure is used to model the nonlinear, time-varying background noise. The noise is removed by subtracting the predicted noise from the original signal. Preliminary experimental results indicate that the ANN based approach consistently outperforms the conventional bandpass filtering approach and the linear adaptive filtering approach. Such performance enhancement is most critical toward the development of a practical automated online ECG arrhythmia monitoring system.<>