Lightweight Denoising Autoencoder Design for Noise Removal in Electrocardiography

Abstract

This study proposes two denoising autoencoder models with discrete cosine transform and discrete wavelet transform, to remove electrode motion artifacts in noisy electrocardiography. Initially, the discrete cosine transform and discrete wavelet transform efficiently removed the high-frequency noise. The six encoder layers then retain important electrocardiography features, whereas the six decoder layers reconstruct the clean electrocardiography. To improve the denoising performance, two network layers, the residual block and pixel adjustment, are added to the encoder and decoder layers to solve the vanishing gradient and improve subtle feature extraction. The proposed methods were applied to 66,000 real-recorded noisy electrocardiography fragments. The experimental result indicates that discrete wavelet transform based denoising autoencoder and discrete cosine transform based denoising autoencoder can improve the signal-to-noise ratio by 25.29 and 25.13 dB on average when the input signal-to-noise ratio is −6 dB.

Keywords

• Artificial neural networks
• biomedical computing
• biomedical signal processing
• discrete cosine transforms
• deep learning
• discrete wavelet transforms