| Summary: | Diabetes mellitus is a disease with significant impact in public health. It is a complex disorder of carbohydrate, fat and protein metabolism that is a result of a deficiency, or complete lack of insulin secretion by the Beta cells of pancreas, or resistance to Insulin. There are 3 types of diabetes, namely type 1 where the patient is insulin-dependent, type 2 where the patient is non insulin-dependent and gestational diabetes that appears during the pregnancy phase. Retinopathy is a diabetes complication that can result in blindness. If detected in an early stage, it can be treated by laser surgery. However its early detection is frequently missed, since it progresses without symptoms until irreversible vision loss occurs. So if we can detect/find cotton wool spots in eye fundus scope by using image recognition, automatic annotation, decision-support systems for risk assessment, conjugate with a mobile app acquiring eye fundus images, we might detect early and treat avoiding patient blindness risk. This project aims to develop a smartphone-based on low computational-cost algorithms, which can be highly efficient in the lower quality images of the smartphone camera, that can be used as a decision-support system. This system may also be extended to other eye diseases, as an useful tool for eye health screening in developing countries and enhance the proximity of screening programs to the population. The main expected contribution is to develop a good decision-support system, considering cotton wool spots, together with red dots, instead of the actual system in use in Portugal which only considers red dots. The number of Diabetic Retinopathy cases worldwide justifies the development of an automated decision-support system for quick and cost effective screening of Diabetic Retinopathy.
|
|---|