Comparison of BSS methods for the detection of α-activity components in EEG

Authors:
Sergey Borisov;Alexander Ilin;Ricardo Vigário;Erkki Oja
Affiliations:
Laboratory of Computer and Information Science, Helsinki University of Technology, HUT, Espoo, Finland;Laboratory of Computer and Information Science, Helsinki University of Technology, HUT, Espoo, Finland;Laboratory of Computer and Information Science, Helsinki University of Technology, HUT, Espoo, Finland;Laboratory of Computer and Information Science, Helsinki University of Technology, HUT, Espoo, Finland
Venue:
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Year:
2006

Citing 4
Cited 0

Independent components of magnetoencephalography: localization

Neural Computation
Overlearning in marginal distribution-based ICA: analysis and solutions

The Journal of Machine Learning Research
Denoising Source Separation

The Journal of Machine Learning Research
Nonstationary nature of the brain activity as revealed by EEG/MEG: methodological, practical and conceptual challenges

Signal Processing - Neuronal coordination in the brain: A signal processing perspective

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we tested the efficiency of a two-step blind source separation (BSS) approach for the extraction of independent sources of α-activity from ongoing electroencephalograms (EEG). The method starts with a denoising source separation (DSS) of the recordings, and is followed by either an independent component analysis (ICA) or a temporal decorrelation algorithm (FastICA and TDSEP, respectively). This two-step method was compared with DSS, ICA and TDSEP alone. The tests were performed with simulated data based on real EEG signal, to guarantee the existence of a “ground truth”. The most efficient algorithm, for proper component extraction (regardless of the amount of α-activity in their spectra) is a combination of DSS and ICA. It provided also more stable results than ICA alone. TDSEP, in combination with DSS, was efficient only for the extraction of the components with prominent α-activity.