Caracterização do eletroencefalograma normal em situação de vigília: elaboração da base de dados e análise quantitativa

• Ano de defesa: 2017
• Autor(a) principal: Ramos, Camila Davi
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Uberlândia; Brasil; Programa de Pós-graduação em Engenharia Elétrica
• Palavras-chave em Português: Engenharia elétrica; Eletroencefalografia; Banco de dados; Processamento de sinais - Técnicas digitais; Quantificação; Processamento digital de sinais; Coma; Densidade espectral de potência; Frequência Mediana; Coerência; Delta; Teta; Alfa; Beta; Gama; Supergama; Ruído; Electroencephalography; Quantification; Digital signal processing; Comatose; Database; Power Spectral Density; Median Frequency; Coherence; Theta; Alpha; Gamma; Noise; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA
• Link de acesso: https://repositorio.ufu.br/handle/123456789/19571; http://doi.org/10.14393/ufu.di.2017.574

The EEG signal, whose information describes the electrical behavior of the cerebral cortex, although it is widely used for diagnoses, mainly of pathologies such as epilepsy, in Brazil it is still not usual to monitor the ICU environment in public hospitals. Considering this, and assuming that studies on normal EEG, registered in people without neurological problems, are scarce, the creation of a base of normal EEG registers and quantitative analysis of it is necessary so that, through the obtained results, Normal patterns can be established and through them, the identification of pathological parameters becomes more effective. In this project, an EEG database was developed, with 100 valid records from normal and healthy volunteers. In addition, from these records, the waking and closed eyes situation was analyzed under the aspect of three distinct quantifiers, being: Power Contribution Percentage (PCP), Median Frequency (FM) and Coherence, both evaluating the signal in the frequency domain. In order to obtain comparisons for the results obtained by the analysis of the normal EEG data, 128 EEG records were used in coma, with different types of etiologies and outcomes. The rhythms that presented the highest distinctions between normal and coma were Delta and Alpha, mainly for the FM quantifier. It was noted that PCP evaluated power characteristics and therefore synthesized the energy information of each brain rhythm in both normal EEG and EEG coma. Already FM brings information of values of frequencies in which there is greater concentration of power, and finally the quantifier coherence informs the degree of similarity between the right and left hemisphere of the brain. Thus, it was not possible to say which of the quantifiers presented better results, since each one is a distinct characterization.