Abstract
Abstract Noise level estimation plays an important role in many applications of signal and image processing, like denoising, compression and detection. Recently, deep neural networks have also been increasingly used for this purpose. In this paper, we develop an effective algorithm of noise level estimation of ECG segments based on trained denoising autoencoder (DAE) with a statistical thresholding method. An important observation is that a well-trained DAE model provides a clean representation of the corrupted training dataset. Two identical cascaded trained DAE models are considered to estimate the statistical properties, e.g., mean and standard deviation, from the trained DAE outputs after applying noise free aligned and jittered training dataset respectively. Two statistical thresholds are calculated from these statistical properties to classify whether the ECG segment is noise-free or jittered or noisy segment. The accuracy of the proposed method is quite promising in classifying and estimating unknow noise level.
Highlights
2 Materials and methodsThe proposed noise level estimation is based on two identical cascaded trained denoising autoencoder (DAE) models (Figure 1)
The proposed DAE model consists of single hidden layer with hyperbolic tangent as activation function [5], where the DAE model is trained with corrupted training and validating dataset of different noise levels (SNR= 5dB, 15dB and 25dB)
Since we trained the DAE model with QRS aligned-noisy training dataset, so the relation between the input and the output of the trained DAE model is directly related to the representation of QRS aligned ECG segments
Summary
The proposed noise level estimation is based on two identical cascaded trained DAE models (Figure 1). They are used to define two thresholds from the statistical properties (e.g., mean μ and standard deviation σ) of their outputs. During the acquisition of biomedical signals, like electrocardiogram (ECG), it is usually corrupted with additive Gaussian white noise. The impact of the noise on the morphology of the ECG signal may be significant.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
