PPGINF PÓS-GRADUAÇÃO EM ENGENHARIA DA INFORMAÇÃO FUNDAÇÃO UNIVERSIDADE FEDERAL DO ABC Téléphone/Extension: Indisponible http://propg.ufabc.edu.br/ppginfo

Banca de DEFESA: ELIANA MARIA DOS SANTOS

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
DISCENTE : ELIANA MARIA DOS SANTOS
DATA : 13/08/2020
HORA: 09:30
LOCAL: por participação remota
TÍTULO:

Improvement of Pre-Processing, Feature Extraction and Classification for Brain-Computer Interfaces in the Motor Imagery Paradigm

 


PÁGINAS: 154
GRANDE ÁREA: Engenharias
ÁREA: Engenharia Biomédica
SUBÁREA: Bioengenharia
ESPECIALIDADE: Processamento de Sinais Biológicos
RESUMO:

Brain-Computer Interfaces (BCI) are used in general to rehabilitate disabled motor functions in the individual, from the modeling of brain signals patterns. One of the paradigms most used in BCI systems is motor imagery (MI), where the acquisition of electroencephalographic (EEG) signals are carried out during execution of tasks requiring concentration on the intention of performing specific body movements, but without actually performing them. The correct identification of brain activity associated to motor imagery is essential for development of functional MI-BCI systems. This study has as main objective the improvement of EEG signal processing for application in MI-BCI, focusing on the pre-processing, feature extraction and classification steps. For this purpose, we investigate several methodologies in each of the mentioned steps, namely: in pre-processing, we investigate a filtering technique in the modulation frequency domain to improve signal-to-noise ratio of EEG signals. The results showed that the proposed solution exceeded the best state-of-the-art results by 3.5%. In the feature extraction step two main methods were used, one based on the Common Spatial Patterns (CSP) and the other based on an EEG signal source localization technique called Low-Resolution Brain Electromagnetic Tomography (LORETA). Using CSP, results were significantly superior to the state-of-the-art results in MI-BCI. However, using LORETA, results were slightly below the results obtained with CSP in the literature. Finally, at the classification stage, a Subject Independent BCI (SI-BCI) model was proposed, which was compared to a traditional Subject Specific BCI (SS-BCI) model. Our results showed that SI-BCI is a viable and feasible option, and its use may be relevant under certain conditions, such as when participants are not available to conduct long training sessions or for a quick evaluation of BCI illiteracy, reducing operating costs and human resources, as well as the stress of subjects involved. In all of the above-mentioned studies, MI public databases available on the internet were used.


MEMBROS DA BANCA:
Presidente - Interno ao Programa - 1545987 - FRANCISCO JOSE FRAGA DA SILVA
Membro Titular - Examinador(a) Interno ao Programa - 1761107 - RICARDO SUYAMA
Membro Titular - Examinador(a) Interno ao Programa - 2334927 - ANDRE KAZUO TAKAHATA
Membro Titular - Examinador(a) Externo ao Programa - 1722875 - DAVID CORREA MARTINS JUNIOR
Membro Titular - Examinador(a) Externo à Instituição - GABRIELA CASTELLANO - UNICAMP
Membro Suplente - Examinador(a) Externo ao Programa - 1955999 - ANDRE MASCIOLI CRAVO
Membro Suplente - Examinador(a) Externo à Instituição - LEVY BOCCATO - UNICAMP
Notícia cadastrada em: 15/06/2020 13:16
SIGAA | UFABC - Núcleo de Tecnologia da Informação - ||||| | Copyright © 2006-2024 - UFRN - sigaa-1.ufabc.int.br.sigaa-1-prod