“Application of deep learning techniques for automating the detection of diabetic retinopathy in retinal fundus photographs”

摘要

“Diabetic Retinopathy” is a disease that is a strong source of blindness in significant cases. It is also found that if it is diagnosed in early stage, it could be handled substantially. Now a days, many tools based on Neural Network and AI are being designed for automatic detection of retinopathy. In this paper, we have done work to increase the accuracy of this automated diagnosis. For this purpose, we have used deep learning technique that detect diabetic retinopathy in patients using their retina fundus photographs. We have achieved an accuracy of 73% which is much more that that of previous work done.