Categorisation and Prognosticationof Diabetic Retinopathy using Ensemble Learning and CNN

Abstract

Diabetic Eye Disease (DED) is a fundamental cause of blindness in the world. As per the various study, 439 million in 2030 and 690 million will be affected in 2045. Different techniques are proposed to forecast and examine the stages in Prognostication of Diabetic retinopathy. The machine learning and deep learning is one of the predominant techniques to project andexplore the images of DR. In this work proposed ensemble-based techniques for classification and prediction of diabetic retinopathy has focused. First the boosting-based ensemble learning method is used to predict the diabetic retinopathy of images. The second the conventional neural network is used to categorise the various stages of the diabetic retinopathy images. The proposed ensemble-based CNN achieved 96 % of accuracy in prophesy of DR images and 98% accuracy in grading of images.