Published in Archives

Is AI better at detecting retinopathy than an ophthalmologist? — Weekly Glance

This is editorially independent content
2 min read

A study published in Ophthalmology Science evaluated general ophthalmologists, retina specialists, and Eyenuk’s EyeArt AI Eye Screening System for detecting diabetic retinopathy (DR).

Tell me about the study.

The study evaluated the sensitivity and specificity of the EyeArt system and dilated eye exams performed by general ophthalmologists and retina specialists against the Early Treatment Diabetic Retinopathy Study (ETDRS) clinical reference standard on the same cohort of 521 study participants. The ETDRS reference standard was established by experts at the University of Wisconsin Reading Center using 10 fundus images per eye captured after dilation by certified photographers, whereas the EyeArt system only analyzed two images per eye, typically without dilation.

What did the study find?

According to the lead author, the sensitivity for the detection of more than mild DR was significantly greater with the EyeArt AI system than a clinical examination by either a general ophthalmologist or a retina specialist. Unlike a few instances in which general ophthalmologists missed some cases of vision-threatening DR, the EyeArt AI system did not miss any cases of vision-threatening DR.

Break these differences down in actual numbers for me.

Sensitivity, a measure of safety (percentage of patients with disease identified correctly), was 96.4% for the EyeArt system in identifying more than mild DR, while that of ophthalmologists’ dilated exams was 27.7% on the identical cohort of study participants. Specificity, a measure of effectiveness (percentage of patients without disease identified correctly), was 99.6% by ophthalmologists’ dilated exams compared to 88.4% with the EyeArt system.

The take home.

This result demonstrates that dilated exams performed by ophthalmologists are better at ruling out disease as evidenced by their high specificity. However, the EyeArt system is much better at identifying patients with disease, a critical factor for a screening scenario in which patients are being identified for referral and further evaluation. According to the authors, the EyeArt system can potentially serve as a low cost point-of-care DR detection tool and help address the diabetic eye screening burden. (via)