Analysis of Diabetic Retinopathy Segmentation and Classifications

EasyChair Preprint 9703

Abstract

In this paper we discussed a framework on the review of Diabetic retinopathy by using various techniques. The retinopathy is mostly observed in diabetic patients. Most of the diabetes faces this complication which leads to vision impairment and blindness among working-age adults. By detecting the stage of DR as early, then they take treatment in time, so, the risk of blindness can be reduced by 95 %.. By using different classifications and segmentations, the stage of retinopathy can be detected as early as possible. In past, there is a method used for evaluating diabetic retinopathy which involves direct and indirect ophthalmoscopy. Segmentation of blood vessels is a main process of color retinal fundus images analysis for diagnosing diabetic retinopathy in ophthalmology. To detect DR in early stage, the system mainly consists of different steps like preprocessing, segmentation, feature extraction and classifications like NPDR and PDR. The DeepDR system is used for detecting the early and late stages of diabetic retinopathy. The segmentation is applied to the images for detection of diabetic retinopathy has some approaches and methodologies. Segmentation, features and classification results of different techniques for various retinopathy images for detecting DR are reviewed.

Keyphrases: Blood vessel segmentation, Classification, DR, Diabetic images, image processing

@booklet{EasyChair:9703,
  author    = {Rayudu Prasanthi and Rajyalakshmi Uppada and B.Leela Kumari},
  title     = {Analysis of Diabetic Retinopathy Segmentation and Classifications},
  howpublished = {EasyChair Preprint 9703},
  year      = {EasyChair, 2023}}