Análise do nível de estresse baseada em sinais de eletroencefalografia e de condutância da pele

• Autor(a) principal: Pomer-Escher, Alexandre Geraldo
• Data de Publicação: 2015
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da Universidade Federal do Espírito Santo (riUfes)
• Texto Completo: http://repositorio.ufes.br/handle/10/1354
• topic: Stress (Fisiologia); Eletroencefalografia; Emoções; Assimetria; Biotecnologia; 61

The stress can affect everyone, independently of age, gender or ethnicity. The human body uses stress as an adaptive response in relations to various conditions, which require an adaptation of the body to face such a situation. Depending on the stressor stimulus, the stress is classified as physical, mental or emotional stress. However, the stress is not necessarily a bad thing or something pathological; in fact, it is a vital adaptation mechanism for the survival of the human species. The number of people who are affected negatively by stress has grown in last decades. Recent research highlights that in the United States about 60% to 90% of medical care were related in some way with stress. In Brazil, around 80% of the population have been affected by stress, and 30% of these are in the most critical phase. Nowadays, the main form of stress identification is still held by self-report questionnaire, thus, this research presents a methodology for stress analysis based on skin conductance and EEG signals. The parameters used for EEG are the asymmetry of alpha rhythm, and the ratio between the beta and alpha rhythms, both in the frontal and prefrontal cortex. For the EEG signal recording handheld device is used. It has electrodes that are specifically located on the positions according to the International 10/20 System (aF3, F3, F4, aF4). The participants are the Military Firefighters of the first Cia of Vitoria (Brazil), and three classes of emotional stimuli are used through the use of images belonging to the IAPS (International Affective Picture System) database, being called positive, calm and negatives classes. The accuracy of results through SVM (Support Vector Machine) classifier are 88.24% for positive stimuli class, 84.09% for calm class and 92.86% for negative class. Thus, this research presents a combination of parameters that can be measured with low-cost equipment, and provides conditions to differentiate stressful stimuli, and thereby providing indications that can be used to help the training of emergency care professionals.