MULTIPLE EYE DISEASE DETECTION USING CONVOLUTIONAL NEURAL NETWORK

Abstract

ABSTRACT: Among the most important systems in the body is the eyes.  Although  their small stature, humans are unable to imagine existence without it. The human  optic is safe against dust particles by a narrow layer called the conjunctiva. It prevents  friction  during the  opening  and  shutting  of  the  eye  by  acting  as  a lubricant.  A  cataract  is  an  opacification  of  the  eye's  lens.  There are  various forms of eye problems. Because the visual system is the most important of the four sensory organs, external eye abnormalities must be detected early. The classification technique can be used in a variety of situations. To examine multiple eye disease detection algorithm based on Optical Coherence Tomography (OCT) scans using Convolution Neural Network (CNN). The proposed work has been deployed with convolution neural networks performed on the OCT image from authenticated data set and accuracy of 91% was achieved using 5-fold cross validation. The highest AUC value for the normal class observed as 1. More than 90% AUC in predicting eye disease for all the classes has been attained using the proposed approach.