Current Trends, Challenges, and Future Prospects for automated detection of Diabetic Retinopathy

Diabetes is a chronic disease which denotes the presence of high levels of glucose in the blood. As of 2019, 77 Million people in India are assumed to be suffering from Diabetes making this count second largest in the world. This number is said to increase up to 134 Million in next 25 years. Diabetes often leads to a chronic eye-disease known as Diabetic Retinopathy. Diabetic retinopathy is a serious ailment and if left undiagnosed, can lead to total loss of vision. Diabetic retinopathy does not have many symptoms thus, a diluted eye examination is essential for its detection. Thus, several studies have been done with the help of Machine Learning and Deep Learning models to support the existing healthcare system for the detection of diabetic retinopathy. In this paper, a comparative study has been done between existing diabetic retinopathy detection and classification methods to provide clearer insights and understanding in this field for future research work.