Multi Label Classification Of Retinal Disease On Fundus Images Using AlexNet And VGG16 Architectures

Abstract

Diseases of the eye have the potential to cause blindness in sufferers. There have been many types of diseases that exist in the human eye. Some examples of diseases that exist in the eye include Diabetic Retinopathy (DR), Myopia (MA), Optic Disc Cupping (ODC). Fundus images help medical personnel to see what diseases are in the eyes of people with certain diseases. In one fundus image there may be more than one disease in the eye. The research that will be carried out is to find out what diseases are contained in the fundus image by using multi-label classification. The research will be conducted using a deep learning method using the AlexNet and VGG16 architectures which will then be compared between the two models. The data used are fundus images on DR, MA, and ODC diseases as many as 1133 data. The results obtained in this study indicate that the AlexNet model is better than the VGG16 model in performing multi-label classification on fundus images.