Desenvolvimento de um dispositivo SSVEP rápido e confiável utilizando eletrodos a seco e frequências acima de 25 Hz

This paper presents a new approach for the processing and classification of visual evoked potentials of steady state (SSVEP). It introduces a ensemble tree model that combines canonical correlation analysis data with methods based on estimation of power spectral density. The stimuli were created usi...

Descripción completa

Detalles Bibliográficos
Autor: Silva, Andrei Damian da
Tipo de recurso: tesis de maestría
Estado:Versión publicada
Fecha de publicación:2018
País:Brasil
Institución:Universidade Federal de Goiás (UFG)
Repositorio:Repositório Institucional da UFG
Idioma:portugués
OAI Identifier:oai:repositorio.bc.ufg.br:tede/8215
Acceso en línea:http://repositorio.bc.ufg.br/tede/handle/tede/8215
Access Level:acceso abierto
Palabra clave:EEG
ICC
Potenciais evocados visuais
SSVEP
BCI
Visual evoked potentials
CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO
Descripción
Sumario:This paper presents a new approach for the processing and classification of visual evoked potentials of steady state (SSVEP). It introduces a ensemble tree model that combines canonical correlation analysis data with methods based on estimation of power spectral density. The stimuli were created using LEDs, from 7.04 Hz to 38.46 Hz. Data were collected using the Texas Instruments ADS1299EEG-Fe and three electrodes. The tests were performed for different distances and light intensities to evaluate the performance of the algorithm under different conditions. In all, 22 participants were recruited, and the average classification was 99.1 ± 2.27% with fixed decision time of 1 second.