Authors:
Andriy Ivannikov
;
Tommi Kärkkäinen
;
Tapani Ristaniemi
and
Heikki Lyytinen
Affiliation:
University of Jyväskylä, Finland
Keyword(s):
Convergence, EEG, ERP, Denoising, Weighted Averaging, SNR, Source Separation.
Related
Ontology
Subjects/Areas/Topics:
Applications and Services
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computer Vision, Visualization and Computer Graphics
;
Medical Image Detection, Acquisition, Analysis and Processing
Abstract:
In this article we compare the convergence rates at increase of the number of processed trials of the three methods applied nowadays in electroencephalography research to denoising of event-related potentials: traditional averaging, weighted averaging, and ERPSUB. We derive the weighted averaging procedure by maximizing
signal-to-noise ratio in the averaged subject responses and show, thereby, that maximizing signal-to-noise ratio criterion is equivalent to minimizing the originally proposed mean-square error criterion in the sense of the weighted averaging problem solving. Moreover, in order to characterize fully the performance of the selected methods, we compare also noise reduction rates in estimates of event-related potentials provided by methods, while the number of processed trials increases.