Detalhes bibliográficos
Ano de defesa: |
2019 |
Autor(a) principal: |
SILVA, Italo Francyles Santos da
![lattes](/bdtd/themes/bdtd/images/lattes.gif?_=1676566308) |
Orientador(a): |
ALMEIDA, João Dallyson Sousa de
![lattes](/bdtd/themes/bdtd/images/lattes.gif?_=1676566308) |
Banca de defesa: |
ALMEIDA, João Dallyson Sousa de
,
BRAZ JÚNIOR, Geraldo
,
TEIXEIRA, Jorge Antônio Meireles
,
ARAÚJO, Sidnei Alves de
![lattes](/bdtd/themes/bdtd/images/lattes.gif?_=1676566308) |
Tipo de documento: |
Dissertação
|
Tipo de acesso: |
Acesso aberto |
Idioma: |
por |
Instituição de defesa: |
Universidade Federal do Maranhão
|
Programa de Pós-Graduação: |
PROGRAMA DE PÓS-GRADUAÇÃO EM CIÊNCIA DA COMPUTAÇÃO/CCET
|
Departamento: |
DEPARTAMENTO DE INFORMÁTICA/CCET
|
País: |
Brasil
|
Palavras-chave em Português: |
|
Palavras-chave em Inglês: |
|
Área do conhecimento CNPq: |
|
Link de acesso: |
https://tedebc.ufma.br/jspui/handle/tede/2592
|
Resumo: |
According to Brazilian Council of Ophthalmology - CBO (2014), 4 million people have some visual impairment. 33 thousand children are blinded by diseases that could be avoided or treated in time. Brückner Test interests to public health. Also known as red-reflex examination, it is an important way for early diagnosis and prevention of optical diseases. It may bring a positive social impact. In this context, this work proposes an automatic method for optical pathologies detection in Brückner test images. This method is based on image processing and machine learning algorithms, contributing for deployment of an accurate computer-aided diagnosis system that helps to avoid and prevent diseases. The proposed method uses texture and color analysis techniques and machine learning to classify cases in healthy or unhealthy. The proposed method reaches 95.25% accuracy, 84.66% sensibility, and 98.90% specificity by using Support Vector Machine classifier. |