Enhancing EEG spacial resolution using ESI: a motor control study

• Ano de defesa: 2019
• Autor(a) principal: Schmiele, Eric Ferreira
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Biomédica
• Palavras-chave em Português: Electroencephalography; Spatial resolution; Electromagnetic Source Imaging; Motion control; Single limb; CNPQ::ENGENHARIAS::ENGENHARIA BIOMEDICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/28620; http://doi.org/10.14393/ufu.di.2020.40

Though electroencephalography (EEG) signals have been widely used to extract brain information they have a low spatial resolution and are not as capable as invasive techniques for applications which demand more specific information from the brain, which would highly enhance the power of brain machine interfaces (BMIs). Because of that, there is a great interest in enhancing EEG's spatial resolution using electromagnetic source imaging (ESI), which combines EEG signals and magnetic resonance images (MRI) to reconstruct the brain's internal current sources (CSs). To investigate the level of information enhancement that ESI can achieve, we proposed to investigate the brain activity differences in signal behavior and position during different movements from the same limb. To do that we applied ESI to simulated data for initial validation, reaching an average correlation coefficient of 0.99 and an average physical displacement of 15.0 mm comparing simulated and calculated signals. Then we applied ESI to EEG recordings of hand, wrist and elbow movements and compared the source signals to the expected neural behavior known from the literature. Thanks to ESI, we reconstructed more than 8000 source points from a total of 61 electrodes enhancing the resolution from an average of 24.69 mm to 3.67 mm. Thanks to this enhancement we were able to analyze spatial and time-frequency information which correlated with the performed movements in accordance to the literature. Therefore, we showed the potential of ESI for applications that demand a better interpretation of brain signals in questions of control of different movements from a single limb.