Hybridization Of Optimized Support Vector Machine And Artificial Neural Network For The Diabetic Retinopathy Classification Problem
Loading...
Date
2019-03
Authors
Ab Kader, Nur Izzati
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
Diabetic Retinopathy (DR) is one of the most threatening disease which caused blindness
for diabetic patient. With the increasing number of DR cases nowadays, diabetic eye
screening has become a challenging task for ophthalmologist as they need to deal with a large
number of retinal image to be diagnosed every day. Screening and early detection of DR play
a vital role to help reducing the incidence of visual morbidity and vision loss. The screening
task is done manually in most countries using qualitative scale to detect abnormalities on the
retina. Although this approach is useful, the detection is not accurate. Previous researchers
have tried a few attempts to propose an automatic DR classification, however it needs to be
improvised especially in terms of accuracy. A group of literates showed that DR classification
can be performed using the clinical features resulted from the blood test such as glycated
haemoglobin, triglyceride, creatine and glucose value. Even this subject have been studied
previously, but it remains the subject of on-going research. Hence, this research aims to obtain
optimal or near-optimal performance value in the study of diabetic classification using
supervised machine learning.
Description
Keywords
Artificial Neural Network , Diabetic Retinopathy Classification