deepthiir/Detection-of-Diabetic-Retinopathy-using-Convolution-Neural-Network

Diabetic Retinopathy (DR) is an eye disease classified on a scale of 0-4. My task was to create an automated analysis system capable of assigning a score based on this scale • High resolution retina images were available which were cropped, resized, converted to 50% grayscale using OpenCV and pandas. Data augmentation methods were used to increase training set • Various convolutional neural networks (CNN) having different number of hidden layers were trained using tensorflow • The best CNN achived the training accuracy of 83.80% and test accuracy of 79.34%