DIABETIC RETINOPATHY DETECTION USING MOBILENETV2 MODEL

Abstract

Diabetic Retinopathy is caused by damage to the blood vessels In the tissue at the back of the eye(Retina).DR is a common complication of diabetes mellitus,which causes lesions on the retina that effect vision.If it is not detected early it can lead to blindness.DR early detection and treatment can significantly reduce the risk of vision loss.Unlike computer-aided diagnosis systems, opthalmologists must manually diagnose DR retina fundus images, which takes time, effort, and money and is prone to error. Convolutional neural networks are becoming more popular as a deep learningmethod for image analysis, and they are extremely effective.We implemented the MobileNetV2 model in CNN to Detect that the image contains which type of DR.Inspite of the stages of DR we implemented that the image contains DR or NO DR separatly.From the results of the Experiments,the highest Accuracy Values are 75% for five stages and 95% for two stages.Two stage detection produced a Precision score of 94%,Recall score of 97%. Key Words: Diabetic Retinopathy, Retina Fundus Images, CNN, MobileNetV2.