HMN 2026: How AI-OCT-based system is noninferior for secondary diabetic macular edema detection

Incorporation of an artificial intelligence-based-optical coherence tomography (AI-OCT) system as a secondary screening tool is noninferior to standard practice and can reduce unnecessary diabetic macular edema (DME) referrals, according to a study published online June 15 in the Journal of the American Medical Association.

Shuyi Zhang, Ph.D., from The Chinese University of Hong Kong, and colleagues examined the diagnostic and referral performance of an AI-OCT system for DME detection within a diabetic retinopathy screening pathway in clinical settings in a prospective silent-mode validation involving 603 patients with diabetes, followed by a multicenter noninferiority randomized controlled trial (RCT) recruiting 276 patients with suspected DME.

RCT participants were randomly assigned to intervention (referral for DME evaluation based on fundus photograph-based screening reports and AI-OCT reports) or control (automatic referral based solely on fundus photograph-based screening reports; 137 and 139 participants, respectively).

The researchers found that the system achieved 98.8% and 90.7% sensitivity and specificity, respectively, for DME detection in the prospective silent-mode validation. DME prevalence was similar in the intervention and control groups in the RCT (30.9% versus 29.9%). The false-positive DME referral rate was 24.1% and 69.1%, respectively (P

“This finding provides a practical framework for the real-world implementation of AI-enabled tools in ophthalmology and other clinical specialties,” the authors write.

The content is provided for information purposes only.