DIABETIC RETINOPATHY DETECTION USING DEEP NEURAL NETWORK

Diabetic Retinopathy (die-uh-BET-ik ret-ih-Nop-uh-thee) is a diabetic complication that effects eyes. It is caused by damage to the blood vessels of the light-sensitive tissues at the back of eye (retina). The condition can developed in anyone who has type 1 or type 2 diabetes. This paper focus on a desktop application that will help you to the identification of diabetic retinopathy. The screening occur in real time. The application can be developed using a tensor flow deep neural network architecture. Here it is trained and tested more than thousands of images. During the creation of deep neural network we will create five layers, 2 pool layer and 2 convolution layer and one fc layer. Fc layers are used to detect specific global configurations of the features detected by the lower layers in the net. In this model there are two options for screening that are one for image screening and one for real time screening. For this desktop application there is no need of internet connection for its working and it can be used as an easy manner.