Detection of Diabetic Retinopathy Using Deep Learning

In this book, a deep learning technique called Convolution Neural Network (CNN) is deployed for determining the level of DR with accuracy. Hence, for training, this network, fundus image dataset from a public resource called Kaggle is utilized. But, before feeding the dataset into a deep learning model, the images are pre-processed using the libraries present in python, so that the model can accurately classify the test input image into the respective diabetic retinopathy class. Thus, the proposed deep learning model helps in identifying DR automatically at its earlier stage, thus helping in controlling its progression that leads to vision-related problems.