25Dec 2023
Real-world evaluation of smartphone-based artificial intelligence to screen for diabetic retinopathy in Dominica

Real-world evaluation of smartphone-based artificial intelligence to screen for diabetic retinopathy in Dominica

A study in Dominica evaluated the performance of an artificial intelligence (AI) system in diabetic retinopathy screening in a real-world setting. Out of 587 participants, the AI demonstrated 77.5% sensitivity and 91.5% specificity in detecting referable diabetic retinopathy, with an overall AUC of 0.8455. Excluding ungradable images, sensitivity and specificity improved to 81.4% and 91.5%, with an AUC of 0.9648. The findings suggest that AI can effectively assist in diabetic screening programs in middle-income settings, offering reasonable accuracy compared to specialist graders.

  • #ophthalmology

Like

Save

Share