Automatic Diabetic Eye Detection Using Machine Learning

The aim of this project is to develop an automated system capable of detecting diabetic retinopathy (DR) in retinal fundus images, which is a common complication of diabetes that can cause vision loss if not identified and treated early. The proposed system utilizes the RESNET 50 algorithm, a deep convolutional neural network, to extract features from retinal images and classify them into one of five DR severity levels. The dataset used in the project comprises 35,126 retinal images The proposed system demonstrated a classification accuracy of 95% in distinguishing diabetic retinopathy, highlighting its potential as an effective tool for early detection and treatment of the condition. Early identification of warning signs of DR is crucial in its management. Hence, regular eye check-ups are necessary to detect the condition and direct the individual to a doctor for prompt examination and treatment, which can help prevent permanent vision loss. However, due to limited resources, screening for DR may not always be possible. To address this challenge, emerging technologies, including artificial intelligence, have been developed for the automatic detection and classification of DR.This study aimed to investigate convolutional neural network used for diagnosing retinopathy.