International Journal of Science and Research (IJSR)

International Journal of Science and Research (IJSR)
Call for Papers | Fully Refereed | Open Access | Double Blind Peer Reviewed

ISSN: 2319-7064

Downloads: 2 | Views: 69

Survey Paper | Biological Engineering | India | Volume 10 Issue 7, July 2021

Survey on Techniques for Diabetic Retinopathy Detection & Classification

Sourabh Naik Khandolkar | Pradnya Mhalsekar | Swapnil Gawade | Pratiksha Shetgaonkar [2] | Amey Gaonkar | Shailendra Aswale [2]

Abstract: Diabetes is a very common disease among people who suffer from glucose abnormalities. Diabetic Retinopathy (DR) is a severe microvascular complication that is found only in diabetic patients that affects the retina of the patient and causes permanent partial or complete blindness to the affected person if not detected and cured at an early stage. However, many diabetic patients fail to detect this disease & lead themselves to visual impairments and this happens due to delay in detection as the patient needs to approach the ophthalmologist who screens the retina of the patient. This manual screening & traditional approach is time-consuming and delays the DR detection process which causes the disease to advance to further stages in the time window and also this manual processing isn?t always accurate. This article surveys & reviews the latest research as well as survey papers discussing about the accurate diagnosis as well as Classification of DR into different categories from mild to severe based on the various techniques utilized for image pre-processing, disease detection & classification from fundus image datasets. In the end the article also proposes the novel model based on ResNet50 (CNN model) for more accurate diagnosis & classification of the DR disease.

Keywords: Machine Learning, Diabetic Retinopathy, Fundus images, Convolutional Neural Network

Edition: Volume 10 Issue 7, July 2021,

Pages: 945 - 952

How to Download this Article?

Type Your Valid Email Address below to Receive the Article PDF Link

Verification Code will appear in 2 Seconds ... Wait