Abnormality Detection and Classification at the Non – Proliferative Stage of Diabetic Retinopathy

EasyChair Preprint 3399

Abstract

Diabetic retinopathy is one of the complications of diabetics, which affects the retinal blood vessels. It is the chronic progressive disease which advances from mild non–proliferative diabetic retinopathy to moderate and to severe proliferative diabetic retinopathy. Early detection and diagnosis can prevent the patient from vision loss. In this work the early stage of reliable segmentation of bright lesions Exudates, is done to detect the severity level .This work has been performed on the standardized online dataset STARE and DRIVE. The main contribution is the reliable vessel extraction and removal which further enhances the quality of feature extraction. The proposed method not only helps for early stage diagnosis but also regarding the treatment decision making.

Keyphrases: Diabetic Retinopathy, Exudates, non-proliferative, vessel extraction

@booklet{EasyChair:3399,
  author    = {S. Pooja and Roy R. U. Renjen},
  title     = {Abnormality Detection and Classification at the Non – Proliferative Stage of Diabetic Retinopathy},
  howpublished = {EasyChair Preprint 3399},
  year      = {EasyChair, 2020}}