Diabetic Retinopathy Classification Using CNN and Hybrid Deep Convolutional Neural Networks

Diabetic Retinopathy (DR) is an eye condition that mainly affects individuals who have diabetes and one of the important causes blindness in adults. As infection progresses, it may lead to permanent loss vision. Diagnosing diabetic retinopathy manually with help ophthalmologist has been a tedious very laborious procedure. This paper not only focuses on detection but also analysis different DR stages, which performed Deep Learning (DL) transfer learning algorithms. CNN, hybrid CNN ResNet, DenseNet are used huge dataset around 3662 train images automatically detect stage progressed. Five 0 (No DR), 1 (Mild 2 (Moderate), 3 (Severe) 4 (Proliferative DR) processed proposed work. The patient’s fed as input model. deep architectures like 2.1 extract features for effective classification. models achieved accuracy 96.22%, 93.18% 75.61% respectively. concludes comparative study highlights perfect classification model automated detection.