Banca de QUALIFICAÇÃO: MARIA EUCLÉCIA ALBUQUERQUE DA SILVA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : MARIA EUCLÉCIA ALBUQUERQUE DA SILVA
DATE: 20/12/2023
LOCAL: Remota
TITLE:

An intelligent system based on electroencephalography to support the diagnosis of autism spectrum disorder

KEY WORDS:

Autism spectrum disorder, diagnostic support systems, autism diagnosis, machine learning, electroencephalography.

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

Autism Spectrum Disorder (ASD) is a nervous system disorder that affects the brain and results in difficulties with speech, social interaction and communication deficits, repetitive behaviors and delays in motor skills. This disorder can usually be distinguished with existing clinical diagnostic protocols from the age of three. ASD also has a genetic influence. One in every 70 children worldwide is affected by ASD. There is no specific treatment for ASD, but several therapeutic techniques have been developed to minimize symptoms and improve cognitive abilities, social and communicative skills and overall quality of life. Currently, the diagnosis of ASD is clinical: teams of psychiatrists, clinical psychologists and neuropsychologists, using observations and questionnaires, make the diagnosis and establish the severity range. Several studies have pointed to the possibility of differential diagnoses based on the analysis of electroencephalographic (EEG) signals, to evaluate brain activity through EEG signals. Machine learning techniques have been investigated to build tools to support differential diagnosis. Deep artificial neural networks have proven to be effective in solving complex classification problems, which could greatly assist in the task of supporting diagnosis through multimodal analysis. This project aims to propose a hybrid architecture based on deep artificial neural networks and statistical learning to support the diagnosis of ASD based on the analysis of EEG signals labeled by clinical diagnosis. The architecture will be validated using public databases of EEG signals obtained from real volunteers on and off the autism spectrum.

COMMITTEE MEMBERS:
Interna - 2324068 - CRISTINE MARTINS GOMES DE GUSMAO
Interno - 1807632 - WELLINGTON PINHEIRO DOS SANTOS
Externa à Instituição - JULIANA CARNEIRO GOMES - UFPE