An adaptive filtering approach for electrocardiogram (ECG) signal noise reduction using neural networks

Electrocardiogram (ECG) signals have been widely used in clinical studies to detect heart diseases. However, ECG signals are often contaminated with noise such as baseline drift, electrode motion artifacts, power-line interference, muscle contraction noise, etc. Conventional methods for ECG noise removal do not yield satisfactory results due to the non-stationary nature of the associated noise sources and their spectral overlap with desired ECG signals.In this paper, an adaptive filtering approach based on discrete wavelet transform and artificial neural network is proposed for ECG signal noise reduction. This new approach combines the multi-resolution property of wavelet decomposition and the adaptive learning ability of artificial neural networks, and fits well with ECG signal processing applications. Computer simulation results demonstrate that this proposed approach can successfully remove a wide range of noise with significant improvement on SNR (signal-to-noise ratio).