2014 | OriginalPaper | Buchkapitel
Influence of Signal Preprocessing on ICA-Based EEG Decomposition
verfasst von : Z. Zakeri, S. Assecondi, A. P. Bagshaw, T. N. Arvanitis
Erschienen in: XIII Mediterranean Conference on Medical and Biological Engineering and Computing 2013
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Independent Component Analysis (ICA) has been widely used for analysis of EEG data and separating brain and non-brain sources from the EEG mixture. In this study, we compared decomposition results of the most commonly applied ICA algorithms: AMICA, Extended-Infomax, Infomax and FastICA. We examined 12 conditions of EEG data pre-processing, and assessed the independence and physiological plausibility of the recovered components. The results demonstrate that, in general, there were no significant differences in the decomposition results, while data pre-processing choices had a much more pronounced effect. In conclusion the efficiency of the ICA decompositions is highly dependent on the pre-processing steps applied to the EEG data submitted to ICA, rather than type of ICA applied.