Related Experiment Video
Updated: Jun 8, 2026
07:41
Behavioral Assessment of Visual Function via Optomotor Response and Cognitive Function via Y-Maze in Diabetic Rats
Published on: October 23, 2020
6.0K
Prediction of Diabetes through Retinal Images Using Deep Neural Network
Mahmoud Ragab1,2,3, Abdullah S Al-Malaise Al-Ghamdi4,5,6, Bahjat Fakieh4
1Information Technology Department, Faculty of Computing and Information Technology, King Abdulaziz University, Jeddah 21589, Saudi Arabia.
Computational Intelligence and Neuroscience
|June 13, 2022
Summary
This study introduces a deep learning model using retinal images for early diabetes detection. The convolutional neural network achieved over 95% accuracy, offering a faster, automated screening method.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Diabetic retinopathy and macular edema are microvascular complications of diabetes affecting the retina.
- Manual screening of retinal images for diabetes diagnosis is time-consuming.
Purpose of the Study:
- To develop and evaluate a deep learning model for automated diabetes prediction using retinal images.
- To automate the screening process for diabetic eye diseases.
Main Methods:
- A 7-layer convolutional neural network (CNN) with ReLU activation and MaxPooling was designed.
- Retinal image datasets were preprocessed and normalized for classification.
- The CNN model was trained to classify images as diabetic or non-diabetic.
Main Results:
- The proposed deep neural network achieved a training accuracy exceeding 95%.
- Performance was evaluated using accuracy, precision, recall, and F1 score.
- The model demonstrated superior performance compared to existing state-of-the-art algorithms.
Conclusions:
- Deep learning, specifically CNNs, can effectively automate diabetes prediction from retinal images.
- The developed model shows significant potential for improving early detection and screening of diabetic eye complications.
