Enhanced automatic artifact detection based on independent component analysis and Renyi's entropy

Nadia Mammone, Francesco Carlo Morabito

Research output: Contribution to journalArticlepeer-review

Abstract

Artifacts are disturbances that may occur during signal acquisition and may affect their processing. The aim of this paper is to propose a technique for automatically detecting artifacts from the electroencephalographic (EEG) recordings. In particular, a technique based on both Independent Component Analysis (ICA) to extract artifactual signals and on Renyi's entropy to automatically detect them is presented. This technique is compared to the widely known approach based on ICA and the joint use of kurtosis and Shannon's entropy. The novel processing technique is shown to detect on average 92.6% of the artifactual signals against the average 68.7% of the previous technique on the studied available database. Moreover, Renyi's entropy is shown to be able to detect muscle and very low frequency activity as well as to discriminate them from other kinds of artifacts. In order to achieve an efficient rejection of the artifacts while minimizing the information loss, future efforts will be devoted to the improvement of blind artifact separation from EEG in order to ensure a very efficient isolation of the artifactual activity from any signals deriving from other brain tasks.

Original languageEnglish
Pages (from-to)1029-1040
Number of pages12
JournalNeural Networks
Volume21
Issue number7
DOIs
Publication statusPublished - Sep 2008

Keywords

  • Artifact
  • EEG
  • Independent component analysis
  • Kurtosis
  • Renyi's entropy
  • Shannon's entropy

ASJC Scopus subject areas

  • Artificial Intelligence
  • Neuroscience(all)

Fingerprint Dive into the research topics of 'Enhanced automatic artifact detection based on independent component analysis and Renyi's entropy'. Together they form a unique fingerprint.

Cite this