Fiche de GONZALEZ ASTUDILLO Juliana

Titre : Development of Network Features for Brain-Computer Interfaces
Date : 6 octobre 2022
Heure: 14:00
Résumé : A Brain-Computer Interface (BCI) is a system that can translate brain activity patterns into messages or commands. Recent neuroimaging studies demonstrated that brain connectivity reveals hemisphere lateralization during motor imagery (MI) in BCI. Besides, it is well known that the motor cortex is principally involved in controlling the contralateral side of the body. This work explores the dual contribution of brain network topology and space in modeling MI mental states through functional lateralization. Specifically, I introduce new metrics to quantify segregation and integration within and between the hemispheres, showing that they are highly relevant features for decoding a MI task. In BCI classification, these network properties not only give high accuracy values but also have the advantage of being neurophysiologically interpretable, compared to state-of-the-art approaches that are instead blind to the underlying mechanism.
Lieu : Auditorium de l’Institut du Cerveau (ICM)