Related Experiment Video
Updated: Aug 1, 2026
09:17
Using Retinal Imaging to Study Dementia
Published on: November 6, 2017
A Prospective Study on Diabetic Retinopathy Detection Based on Modify Convolutional Neural Network Using Fundus
Awais Bajwa1, Neelam Nosheen1, Khalid Iqbal Talpur2
1Ophthalytics, Marietta, GA 30062, USA.
Diagnostics (Basel, Switzerland)
|February 11, 2023
Summary
Early detection of Diabetic Retinopathy (DR), a leading cause of blindness, is crucial. A deep learning model achieved high accuracy in classifying DR-positive and DR-negative fundus images, aiding early diagnosis.
Area of Science:
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background:
- Diabetic Retinopathy (DR) is a leading cause of blindness globally, stemming from diabetes.
- Early detection of DR is vital for sight preservation but challenging due to reliance on expert interpretation of fundus images.
Purpose of the Study:
- To develop and validate a deep learning model for automated Diabetic Retinopathy detection.
- To assess the real-time performance of the AI model in a clinical setting.
Main Methods:
- A deep learning model was trained and validated on a private dataset.
- The model was tested in real-time at the Sindh Institute of Ophthalmology & Visual Sciences (SIOVS), evaluating image quality and classifying images as DR-Positive or DR-Negative.
- Clinical experts reviewed the model's classifications.
Main Results:
- The study screened 398 patients over five weeks.
- The deep learning model achieved 93.72% accuracy, 97.30% sensitivity, and 92.90% specificity on test data.
- Model performance was validated against expert clinical labels.
Conclusions:
- The deep learning model demonstrates significant potential for accurate and efficient Diabetic Retinopathy screening.
- Automated DR detection can support clinical experts, improving early diagnosis and preventing vision loss.
