Investigation on Severity Level for Diabetic Maculopathy based on the Location of Lesions

Computerized recognition of lesions in retinal images can support in early diagnosis and screening of a common disease, Diabetic Maculopathy (DM). A computationally well-organized tactic for the localization of lesions in a fundus retinal image is offered in this paper. Diabetic maculopathy is an ophthalmic system disease initiated by complication of diabetes. It is a foremost source of blindness in both middle and advanced age group. Former recognition of diabetic maculopathy defends patient from vision loss. Feature abstraction techniques can shrink the effort of ophthalmologists and are used to gladly perceive the Endurance of aberrations in the retinal images assimilated during the screenings. Lesions are a foremost source of diabetic maculopathy. In this paper the lesions are perceived by means of Fuzzy c means Clustering [FCM] algorithm in the non-dilated retinal images. The developed system consists of image acquisition, image preprocessing with a combination of fuzzy techniques, feature extraction, and image classification. The fuzzy-based image processing decision support system will assist in the diabetic retinopathy screening and reduce the burden borne by the screening team. Feature abstraction dramas a energetic protagonist in perceiving the diseases. And also it concentrates on ruling the brutality of disease through Cascade Neural Network (CNN) Classifier and Fuzzy Classifier. Citation: P. Bhuvaneswari, R. Banumathi. "Investigation on Severity Level for Diabetic Maculopathy based on the Location of Lesions." Global Research and Development Journal For Engineering 4.7 (2019): 39 - 47.