Decomposition Methods for Detailed Analysis of Content in ERP Recordings

Authors:
Vasiliki Iordanidou;Kostas Michalopoulos;Vangelis Sakkalis;Michalis Zervakis
Affiliations:
Department of Electronic and Computer Engineering, Technical University of Crete, Chania, Greece 73100;Department of Electronic and Computer Engineering, Technical University of Crete, Chania, Greece 73100;Institute of Computer Science, Foundation for Research and Technology, Heraklion, Greece 71110;Department of Electronic and Computer Engineering, Technical University of Crete, Chania, Greece 73100
Venue:
ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part II
Year:
2009

Citing 1
Cited 1

An information-maximization approach to blind separation and blind deconvolution

Neural Computation

Time-frequency data reduction for event related potentials: combining principal component analysis and matching pursuit

EURASIP Journal on Advances in Signal Processing - Special issue on applications of time-frequency signal processing in wireless communications and bioengineering

Quantified Score

Hi-index	0.00

Visualization

Abstract

The processes giving rise to an event related potential engage several evoked and induced oscillatory components, which reflect phase or non-phase locked activity throughout the multiple trials. The separation and identification of such components could not only serve diagnostic purposes, but also facilitate the design of brain-computer interface systems. However, the effective analysis of components is hindered by many factors including the complexity of the EEG signal and its variation over the trials. In this paper we study several measures for the identification of the nature of independent components and address the means for efficient decomposition of the rich information content embedded in the multi-channel EEG recordings associated with the multiple trials of an event-related experiment. The efficiency of the proposed methodology is demonstrated through simulated and real experiments.