Real-world artificial intelligence-based interpretation of fundus imaging as part of an eyewear prescription renewal protocol
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.Ophthalmic technicians can use AI software for sensitive screening of fundus abnormalities during French eyewear renewal. The AI demonstrated 100% sensitivity and 94.4% specificity in detecting eye conditions.
Area Of Science
- Ophthalmology
- Medical Imaging
- Artificial Intelligence
Background
- The French eyewear prescription renewal protocol (RNO) provides an opportunity for routine fundus examinations.
- Artificial intelligence (AI) offers potential for automated detection of fundus abnormalities.
Purpose Of The Study
- To evaluate the real-world diagnostic accuracy of the Opthai® AI software for detecting fundus abnormalities.
- To assess the software's performance within the RNO protocol context.
Main Methods
- A retrospective, single-center review of 2056 fundus images from 1028 patients.
- Comparison of AI software (index test) operated by technicians against ophthalmologist diagnoses (reference test).
Main Results
- The AI software detected abnormalities in 7.2% of images, while ophthalmologists confirmed them in 1.7%.
- The AI achieved 100% sensitivity and 94.4% specificity.
- False positives were mainly glaucoma suspects; OCT imaging facilitated diagnosis for 81% of these cases.
Conclusions
- AI software is suitable for ophthalmic technicians to perform highly sensitive screening of fundus abnormalities.
- This approach can identify patients requiring ophthalmologist evaluation efficiently within the RNO protocol.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.