Classification of retinal images with deep learning for early detection of diabetic retinopathy disease
Problem
Diabetic retinopathy, a leading cause of blindness in people with diabetes, needs to be detected early from retinal images to prevent vision loss.
Why it matters
Early detection is critical because the disease is the most common cause of diabetes-related blindness.
What the AI did
A Convolutional Neural Network automatically learned features from preprocessed, standardized retinal images and classified them, achieving 98.5% success without manual feature engineering.
Results
98.5% classification success stated; dataset and metric details not stated in source
Impact
Demonstrated automated retinal-image classification in a study setting; no clinical deployment or external validation stated.
Record
| Venue | Signal Processing and Communications Applications Conference (SIU) |
|---|---|
| Published | 2018 |
| Work type | conference-paper |
| Authors | Nursel Yalçın, Seyfullah Alver, Necla Uluhatun |
| Citations | 43 (OpenAlex, at import time) |
| References | 7 |
| Model or method | Convolutional Neural Network (CNN) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1109/siu.2018.8404369 |
| Open access | closed |
| Keywords | diabetic retinopathy, deep learning, CNN, retinal images, classification |
Importance rating
significant — thousands to millions affected
Abstract states diabetic retinopathy is the most common cause of diabetes-related blindness and early detection is critical; a major vision-threatening condition, though study is a research-stage classifier.