A comprehensive approach for retinal image analysis for the detection of diabetic retinopathy

Abstract

Diabetic Retinopathy (DR) is a common diabetic complication that occurs when the retinaand#8223;s blood vessels are damaged due to high blood sugar levels. This results in swelling and leaking of the vessels. As DR progress, diabetic patients may experience Diabetic Macular Edema (DME). The accretion of fluid inside the macula due to the disordered blood retina results in DME. By 2025, it is projected to be around 520 million people can have diabetes, which counts around 382 million in 2013. According to approximations, 10% of those who have diabetes are at the threat of DME. There are currently 26.7 million DME patients worldwide. It is one of the main causes of vision impairment and if left untreated, can possibly result in blindness. Therefore, early detection is necessary to solve this vision issue. The manual diagnosis of DR requires a lot of time and is susceptible to disease misinterpretation. There exists several segmentation and classification methods related to DR detection, many of them fall short in their ability to extract the blood vessel and abnormalities accurately and efficiently from every single medical image. This can be enhanced by implementing an automated system formulated using optimization techniques. newline