An MLP neural network for ecg noise removal based on kalman filter

Sara Moein

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

In this paper, application of Artificial Neural Network (ANN) for electrocardiogram (ECG) signal noise removal has been investigated. First, 100 number of ECG signals are selected from Physikalisch-Technische Bundesanstalt (PTB) database and Kalman filter is applied to remove their low pass noise. Then a suitable dataset based on denoised ECG signal is configured and used to a Multilayer Perceptron (MLP) neural network to be trained. Finally, results and experiences are discussed and the effect of changing different parameters for MLP training is shown.

Original languageEnglish
Title of host publicationAdvances in Computational Biology
EditorsHamid Arabnia
Pages109-116
Number of pages8
DOIs
StatePublished - 2010
Externally publishedYes

Publication series

NameAdvances in Experimental Medicine and Biology
Volume680
ISSN (Print)0065-2598

Keywords

  • Dataset
  • Kalman filter
  • MLP training
  • Noise removal
  • Performance

Fingerprint

Dive into the research topics of 'An MLP neural network for ecg noise removal based on kalman filter'. Together they form a unique fingerprint.

Cite this