Diabetic Retinopathy Detection Using Transfer Learning

Abstract

The proposed research work deals with the image processing and machine learning algorithms to examine fundus images for the diagnosis and categorization of diabetic retinopathy, which affects a substantial number of humans worldwide due to diabetes. Pre-processing and feature extraction from the extracted images will be part of the research, along with training and assessing of deep learning models for categorization technique. The proposed model will predict the stage of diabetic retinopathy from fundus retinal images. The research carried out provides an output stages of diabetic conditions in a patient, such as: No Diabetic Retinopathy, Mild, Moderate, Severe, and Proliferative. The ultimate aim of the model is to develop a tool that will help medical practitioners diagnose using technologies such as artificial intelligence or internet of things, etc. for treating diabetic retinopathy to improve efficient outcomes on diagnosis.