Abstract
Recently, deep learning models such as deep convolutional neural networks have shown great success in different imaging tasks in general and in detection of diabetic retinopathy in particular. An automated detection system is undoubtedly a great help for the screening stage of diabetic retinopathy for further treatment to reduce the burden on the public health system. In this paper, a general framework for automated detection virtual environment is developed as a virtual hospital using a convolutional neural network (CNN) approach as imaging end. Image classification is done in two stages. The best model being 87.12% accurate for the first stage on more than 53000 images which is better than previous works tested only on very less amount of test data. Using this model, a standalone application is generated with extra features for user interaction to be used by the patient in the virtual hospital when he presses a key on the keyboard.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call