Publicación:
A Novel EEG-Based Four-Class Linguistic BCI
A Novel EEG-Based Four-Class Linguistic BCI
No hay miniatura disponible
Fecha
2019
Autores
Jahangiri, Amir
Achanccaray, David
Sepulveda, Francisco
Título de la revista
Revista ISSN
Título del volumen
Editor
IEEE
Proyectos de investigación
Unidades organizativas
Número de la revista
Abstracto
In this work, we present a novel EEG-based Linguistic BCI, which uses the four phonemic structures "BA", "FO", "LE", and "RY" as covert speech task classes. Six neurologically healthy volunteers with the age range of 19-37 participated in this experiment. Participants were asked to covertly speak a phonemic structure when they heard an auditory cue. EEG was recorded with 64 electrodes at 2048 samples/s. The duration of each trial is 312ms starting with the cue. The BCI was trained using a mixed randomized recording run containing 15 trials per class. The BCI is tested by playing a simple game of "Wack a mole" containing 5 trials per class presented in random order. The average classification accuracy for the 6 users is 82.5%. The most valuable features emerge after Auditory cue recognition (~100ms post onset), and within the 70-128 Hz frequency range. The most significant identified brain regions were the Prefrontal Cortex (linked to stimulus driven executive control), Wernicke's area (linked to Phonological code retrieval), the right IFG, and Broca's area (linked to syllabification). In this work, we have only scratched the surface of using Linguistic tasks for BCIs and the potential for creating much more capable systems in the future using this approach exists.
Descripción
Palabras clave
Training,
Task analysis,
Electroencephalography,
Time-frequency analysis,
Phonetics,
Protocols