Riemannian geometry for combining functional connectivity metrics and covariance in BCI - Laboratoire Interdisciplinaire des Sciences du Numérique Access content directly
Journal Articles Software Impacts Year : 2022

Riemannian geometry for combining functional connectivity metrics and covariance in BCI

Abstract

Brain–computer interfaces allow interactions based on brain activities detected in electroencephalography. Despite important improvements in the last decade, some subjects still achieve poor performances without any identified cause. On the one hand, State-of-the-art methods for online decoding are based on covariance matrices seen as elements of a Riemann manifold. On the other hand, functional connectivity is a powerful method to characterize the brain activity. The proposed software combines functional connectivity and covariance within a Riemannian framework to increase the robustness of brain–computer interfaces.
Fichier principal
Vignette du fichier
1-s2.0-S2665963822000197-main.pdf (695.84 Ko) Télécharger le fichier
Origin : Publisher files allowed on an open archive

Dates and versions

hal-03772666 , version 1 (21-09-2022)

Licence

Attribution

Identifiers

Cite

Sylvain Chevallier, Marie Constance Corsi, Florian Yger, Fabrizio de Vico Fallani. Riemannian geometry for combining functional connectivity metrics and covariance in BCI. Software Impacts, 2022, 12, pp.100254. ⟨10.1016/j.simpa.2022.100254⟩. ⟨hal-03772666⟩
123 View
65 Download

Altmetric

Share

Gmail Facebook X LinkedIn More