Recent works have shown that artifact removal in biomedical signals can be performed by using Discrete Wavelet Transform (DWT) or Independent Component Analysis (ICA). It results often very difficult to remove some artifacts because they could be superimposed on the recordings and they could corrupt the signals in the frequency domain. The two conditions could compromise the performance of both DWT and ICA methods. In this study we show that if the two methods are jointly implemented, it is possible to improve the performances for the artifact rejection procedure. We discuss in detail the new method and we also show how this method provides advantages with respect to DWT of ICA procedure. Finally, we tested the new approach on real data.

Independent Component Analysis and Discrete Wavelet Transform for Artifact Removal in Biomedical Signal Processing / Calcagno, Salvatore; LA FORESTA, Fabio; Versaci, Mario. - In: AMERICAN JOURNAL OF APPLIED SCIENCES. - ISSN 1546-9239. - 11 (1):1(2014), pp. 57-68. [10.3844/ajassp.2014.57.68]

Independent Component Analysis and Discrete Wavelet Transform for Artifact Removal in Biomedical Signal Processing

CALCAGNO, SALVATORE;LA FORESTA, Fabio;VERSACI, Mario
2014-01-01

Abstract

Recent works have shown that artifact removal in biomedical signals can be performed by using Discrete Wavelet Transform (DWT) or Independent Component Analysis (ICA). It results often very difficult to remove some artifacts because they could be superimposed on the recordings and they could corrupt the signals in the frequency domain. The two conditions could compromise the performance of both DWT and ICA methods. In this study we show that if the two methods are jointly implemented, it is possible to improve the performances for the artifact rejection procedure. We discuss in detail the new method and we also show how this method provides advantages with respect to DWT of ICA procedure. Finally, we tested the new approach on real data.
2014
Artifact Removal, Discrete Wavelet Transform, Independent Component Analysis, Neural Networks, Surface EMG
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12318/1379
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 21
  • ???jsp.display-item.citation.isi??? ND
social impact