Identification of exudates for Diabetic Retinopathy based on morphological process and PNN classifier

Abstract

Diabetic Retinopathy is an ocular systemic disease caused by complication of diabetes. It is a major cause of blindness in both middle and advanced age group. Earlier detection of diabetic retinopathy protects patient from vision loss. The foremost symptom of this blindness is the exudates. Exudates are the liquefied fluid comprising solutes, proteins, cells, or cellular debris leaked from the damaged blood vessels into nearby tissues or on tissue surfaces in the retina. The leakage of these proteins or lipids causes vision loss to the patients. Identifying the exudates in advance can safeguard the diabetic patients from blindness. Dilation method is used by the ophthalmologists to detect the exudates. But it causes the irritation to the patients' eyes. This paper focuses on an automated method which detects the diabetic retinopathy through detecting exudates by Morphological process in colour fundus retinal images and then segregates the severity of the lesions. The severity level of the disease was achieved by Probabilistic Neural Network (PNN) classifier.