Representing Ethnic Minorities in Chinese Media: A Comparative Study of People’s Daily and Hunan Daily in the Era of Targeted Poverty Alleviation

Abstract

Diabetic Retinopathy (DR) is certainly a common vision disease and substantial reason behind blindness in diabetics. It is one of the problems of diabetes that affects the eyes. If not really at an early on stages, then it can cause long term blindness. DR is a medical condition where the retina is damaged because fluid leaks from blood vessels into the retina. Diabetic retinopathy based on features such as blood vessel, exudes, hemorrhages, micro aneurysms. To accurately train a deep leaning model to classify DR, an enormous number of images is required and this is an important limitation in the DR. In this paper, target task is to implement diabetic retinopathy fundus images classification using CNNs based Transfer learning. Transfer learning technique that can help in overcoming the scarcity of fundus images. The main idea that is exploited by Transfer learning is that deep leaning architecture trained on Non-medical images. This review paper that focus on Automatic DR Detection and classification by using transfer learning to present the best existing methods to address this problem.