Author : Anu Johny and Abraham Thomas

Diabetic macular edema (DME) is an eye disease that belongs to diabetic retinopathy which leads to vision loss. DME, which is a swelling in the macular region of eye is a common vision problem seen in diabetic patients.Hard exudates (bright lesions)in the fundus image indicates the presence of DME. When blood vessels in retina of diabetic patient is begin to leak into the macula region of eye leading to vision loss. Early detection of Macular Edema in diabetic patients prevents blindness. So in this article detection of exudates and its proximity towards macula determines and thereby its everity level. This article presents the computer based approach for the DME detection, which includes the steps image preprocessing, localization and masking of the optic disk, detection of fovea, ROI extraction, motion pattern generation, determination of disease severity level. First stage,detecting of DME and the next stage is assessing the severity of DME. Also this paper presents an automatic method for glaucoma detection.

Keywords: Diabetic Macular Edema, Fundus images, Exudates,ROI, K-Means clustering, Cup to disc ratio(CDR)

Article published in International Journal of Current Engineering and Technology, Vol.5, No.5 (Oct-2015)