Banca de QUALIFICAÇÃO: ANDRESSA LAYSA QUEIROZ RIBEIRO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : ANDRESSA LAYSA QUEIROZ RIBEIRO
DATE: 22/12/2023
LOCAL: Remota
TITLE:

EXPRESSION OF EMOTIONS IN THE ELDERLY: EVALUATION OF FOUR DEEP NEURAL NETWORKS FOR EXTRACTING FEATURES FROM ELECTROENCEPHALOGRAM DATA

KEY WORDS:

data base; EEG; emotions; networks; aging; therapies; elderly.

PAGES: 30
BIG AREA: Engenharias
AREA: Engenharia Biomédica
SUMMARY:

Brazil's demographic distribution has undergone major changes in recent decades, due to an increase in life expectancy that has resulted in a continuous process of population aging. Various socioeconomic, cultural, physiological, environmental and political conditions condition the aging of the population to the prevalence of diseases related to senescence, such as osteoporosis, hypertension and dementia, with Alzheimer's disease being the most common. Dementias are degenerative and progressive conditions that involve memory decline and other cognitive deficits. The challenge that arises is to develop scenarios in which advances in health, science and technology that will allow human beings to achieve quick, efficient and at low cost for the health care of the elderly population, several studies attest that music therapy can slow down the progress of dementia through musical stimuli and musical education that stimulate the areas of the brain responsible for memory through emotions. Recognizing human emotions through computers has been a significant challenge in the field of Computing. To perform emotion recognition, it is necessary to collect human data, the Electroencephalogram (EEG) has been the most used to provide insights into different emotions that are associated with distinct patterns of brain activity and can contribute to early diagnosis, monitoring the progression of the condition , personalization of therapy based on specific patterns of brain activity, evaluation of therapeutic interventions and the development of assistive technologies. This work presents the evaluation of four deep neural networks: Lenet, Resnet, SqueezeNet and VGG, to extract attributes from data collected through EEG that shows potential to support therapy for the elderly, with the possibility of reducing the progress of dementia, including Alzheimer's disease.

COMMITTEE MEMBERS:
Interno - 1807632 - WELLINGTON PINHEIRO DOS SANTOS
Externa ao Programa - 2727505 - GISELLE MACHADO MAGALHAES MORENO - UFPEExterna à Instituição - JULIANA CARNEIRO GOMES - UFPE