Banca de QUALIFICAÇÃO: RAIMUNDO GUIMARAES SARAIVA JUNIOR

Uma banca de QUALIFICAÇÃO de DOUTORADO foi cadastrada pelo programa.
STUDENT : RAIMUNDO GUIMARAES SARAIVA JUNIOR
DATE: 18/04/2023
TIME: 13:00
LOCAL: Sala de Seminários (Espaço Sérgio Barroso)
TITLE:

INTELIGÊNCIA COMPUTACIONAL NO APOIO AO DIAGNÓSTICO DA DOENÇA DE ALZHEIMER

KEY WORDS:

Alzheimer’s disease; machine learning; attribute selection; deep learning

PAGES: 126
BIG AREA: Engenharias
AREA: Engenharia Elétrica
SUBÁREA: Telecomunicações
SPECIALTY: Sistemas deTelecomunicações
SUMMARY:

Since Alzheimer’s disease is complex, difficult to diagnose, and can be mistaken with other forms of dementia, the specific diagnosis of the condition is yet unknown to science. Dementias are becoming more prevalent among the world’s population, which is also living longer, which partly explains the rise in Alzheimer’s disease diagnoses, the most common type of dementia. While neurological deterioration is a normal part of aging, Alzheimer’s disease quickens the process, putting a strain on public health services, causing early mortality, and causing tremendous inconvenience for the patient and his or her family. In this context, this work investigates artificial intelligence techniques that seek to aid in the diagnosis of Alzheimer’s disease, using clinical data and magnetic resonance images. Under licensing for use, primary information was obtained from Australian Imaging, Biomarker & Lifestyle Flagship Study of Ageing (AIBL), Alzheimer’s Disease Neuroimaging Initiative (ADNI), and the Open Access Series of Imaging Studies (OASIS). To treat the clinical data, a new method of attribute selection was proposed, reduced databases were created, and evaluation was performed by machine learning classification models. The use of images obtained from MRI scans was also investigated. Different strategies were proposed to build an image bank suitable for the application of deep learning techniques. Evaluation metrics were computed on the proposed strategies, and diagnostic results were obtained for the cases of normal and established Alzheimer’s individuals. The case of individuals with mild cognitive impairment was also evaluated. As part of the performance evaluation, the achieved results were compared with the results obtained in other research works that are available in the literature.

BANKING MEMBERS:
Interno - 2131191 - DANIEL GUERREIRO E SILVA
Presidente - 1141301 - FRANCISCO ASSIS DE OLIVEIRA NASCIMENTO
Externo à Instituição - HENRIQUE SARMENTO MALVAR - UW
Externo à Instituição - JOSE ANTONIO APOLINARIO JUNIOR - IME
Interno - 1415757 - VINICIUS PEREIRA GONCALVES