Skip to Main content Skip to Navigation
Conference papers

Détection dynamique de graphes appliquée à l'analyse de signaux électroencéphalographiques issus du sommeil de rat

Abstract : The aim of this paper is to propose a method for the estimation of oriented graphs of connectivity between signals from different areas of the brain during different epochs of sleep and wake. The non stationarity is taken into account to evaluate the links between the activities of the different areas to put in evidence the dynamical aspect of the underlying graphs. For that, a method to estimate the partial directed coherence is tested and analysed on a model that generates signals from sequences of multivariate autoregressive models supposed to mimick the dynamical aspect of changes of the links of the areas of the brain during different vigilance states. The method is tested on real data, three electroencephalographic signals coming from three different areas of the brain of a rat recorded during three vigilance states. The method puts in evidence, on one hand, the dynamical behavior of the exchange between the areas of the brain, on the other hand, the distribution of the principal connectivity graphs for the different states of vigilance.
Complete list of metadata

Cited literature [6 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01662574
Contributor : Guillaume Becq <>
Submitted on : Wednesday, December 13, 2017 - 11:46:52 AM
Last modification on : Tuesday, May 11, 2021 - 11:37:37 AM

File

BecqEtAl2017Gretsi326.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01662574, version 1

Citation

Guillaume Becq, Pierre-Olivier Amblard, Paul Salin, Jean-Christophe Comte. Détection dynamique de graphes appliquée à l'analyse de signaux électroencéphalographiques issus du sommeil de rat. GRETSI 2017 - XXVIème Colloque francophone de traitement du signal et des images, Sep 2017, Juan-Les-Pins, France. ⟨hal-01662574⟩

Share

Metrics

Record views

407

Files downloads

214