Généralisation des micro-états EEG par apprentissage régularisé temporellement de dictionnaires topographiques

Le modele des micro-etats decrit les signaux EEG par des suites de topographies associees a des etats cerebraux demeurant stables durant quelques dizaines de millisecondes. La generalisation proposee dans cet article considere un modele redondant autorisant plusieurs etats a etre actifs simultanement. Un apprentissage de dictionnaire regularise temporellement est propose afin d'extraire ces etats. L'efficacite de representation des deux modeles est comparee sur des signaux de synthese ainsi que sur des signaux reels pour l'etude du potentiel evoque P300. Abstract – The microstate model describes EEG signals as series of topographies remaining stable during several tens of milliseconds and associated to brain states. The proposed generalization in this paper is based on an overcomplete model allowing several states to be active simultaneously. A dictionary learning algorithm with a temporal regularization is proposed to extract these states. The representation effectiveness of both models is compared on artificial and real signals for the extraction of the evoked potential P300.

[1]  Klaus-Robert Müller,et al.  The BCI competition 2003: progress and perspectives in detection and discrimination of EEG single trials , 2004, IEEE Transactions on Biomedical Engineering.

[2]  Pascal Frossard,et al.  Dictionary Learning , 2011, IEEE Signal Processing Magazine.

[3]  Dietrich Lehmann,et al.  EEG microstates , 2009, Scholarpedia.

[4]  C. Michel,et al.  Unraveling the cerebral dynamics of mental imagery , 1997, Human brain mapping.

[5]  Guillermo Sapiro,et al.  Online dictionary learning for sparse coding , 2009, ICML '09.

[6]  Dietrich Lehmann,et al.  Brain Electric Microstates and Cognition: The Atoms of Thought , 1990 .

[7]  T. Koenig,et al.  EEG microstate duration and syntax in acute, medication-naïve, first-episode schizophrenia: a multi-center study , 2005, Psychiatry Research: Neuroimaging.

[8]  D. Lehmann,et al.  Larger topographical variance and decreased duration of brain electric microstates in depression , 2005, Journal of Neural Transmission / General Section JNT.

[9]  D Lehmann,et al.  EEG alpha map series: brain micro-states by space-oriented adaptive segmentation. , 1987, Electroencephalography and clinical neurophysiology.

[10]  Jamal Atif,et al.  On the need for metrics in dictionary learning assessment , 2014, 2014 22nd European Signal Processing Conference (EUSIPCO).

[11]  Guillaume Gibert,et al.  “P300 speller” Brain-Computer Interface: Enhancement of P300 evoked potential by spatial filters , 2008, 2008 16th European Signal Processing Conference.

[12]  D. Lehmann,et al.  Segmentation of brain electrical activity into microstates: model estimation and validation , 1995, IEEE Transactions on Biomedical Engineering.

[13]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[14]  Denis Brunet,et al.  Topographic ERP Analyses: A Step-by-Step Tutorial Review , 2008, Brain Topography.

[15]  Cédric Gouy-Pailler,et al.  Multi-dimensional sparse structured signal approximation using split bregman iterations , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.

[16]  Maria Stein,et al.  The early context effect reflects activity in the temporo-prefrontal semantic system: Evidence from electrical neuroimaging of abstract and concrete word reading , 2008, NeuroImage.

[17]  Han Yuan,et al.  Spatiotemporal dynamics of the brain at rest — Exploring EEG microstates as electrophysiological signatures of BOLD resting state networks , 2012, NeuroImage.