Download PDFOpen PDF in browser

Diabetic Retinopathy Detection

EasyChair Preprint no. 7554

5 pagesDate: March 13, 2022


According to the International Diabetes Federation (IDF), the overall population of diabetics in India was around 50.8 million in 2010, and it is expected to climb to 87.0 million by 2030. Diabetic Retinopathy is one of the most common problems associated with Type 2 diabetes. Diabetic Retinopathy is a condition that causes blindness in people aged 20 to 64. Long-term diabetic retinopathy disrupts the normal flow of fluid out of the eye, putting pressure on the eyeball and potentially damaging nerves, which can lead to glaucoma. Diabetic retinopathy can be detected and treated early, which reduces the chance of visual loss.


Manual Diabetic Retinopathy Diagnosis by ophthalmologists involves time, effort, and money, and can lead to misdiagnosis if computer-aided diagnosis systems are not used. Deep Learning has recently emerged as one of the most popular methods for achieving high performance results in a variety of fields, including medical image analysis and classification. This study tackles the topic of predicting diabetic retinopathy in advance in order to avert future consequences. The suggested classifier is based on the Mobile Net architecture, which is a lightweight, mobile-friendly design that was trained on retinal fundus images from the Aptos 2019 challenge data set.

The proposed enhanced model gives an accuracy of 96% and precision, recall, f-1 scores are 0.95, 0.98 and 0.97 respectively. Presented results demonstrate that this model achieves promising results and can be deployed as an application for clinical testing. This work attempts to suggest the diabetic retinopathy complications in advance. The intention of the work is to help the practitioners not to replace the ophthalmologist.

Keyphrases: binary classifier, CNN, deep learning, Diabetic Retinopathy, MobileNetV2

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
  author = {Nischit Patel and Kevin Patel and Kunj Patel and Suvarna Pansambal},
  title = {Diabetic Retinopathy Detection},
  howpublished = {EasyChair Preprint no. 7554},

  year = {EasyChair, 2022}}
Download PDFOpen PDF in browser