Methodology for the Detection of Asymptomatic Diabetic Retinopathy

Diabetic retinopathy, an asymptomatic problem of diabetes, is one of the leading sources of blindness worldwide. The primary detection and diagnosis can decrease the incidence of severe vision loss due to diabetes. Therefore, the present study was conducted to design an experiment in order to diagnose symptomless clinical stages of diabetic retinopathy, i.e., progressive diabetic retinopathy and nonproliferative diabetic retinopathy subjectively and objectively. The diagnostic confirmation of diabetic retinopathy depends on the reliable detection and classification of bright lesions, such as exudates and cotton wool spots, and dark lesions, such as: microan-eurysms and hemorrhages, present in retinal fundus images. However, variations in the retinal fundus images make it difficult to discriminate dark and bright lesions in the existence of landmarks, like blood vessels and optic disk. Thus, it is essential to remove any spurious and false areas caused by anatomical structures before the segmentation of retinal lesions. In addition, to design an efficient computer-aided diagnostic method, a benchmark composite database, having variable characteristics, such as position, dimensions, shapes, and color is required. Keeping all these facts in mind, a composite experimental methodology is designed in this study for an effective analysis of the computer-aided solution for the diagnosis of diabetic retinopathy.