AI Tool May Cut Macular Edema Referrals - Report - MDSpire

AI Tool May Cut Macular Edema Referrals

An artificial intelligence–based optical coherence tomography pathway met noninferiority criteria for false-positive diabetic macular edema referrals and was associated with fewer referral decisions in a randomized clinical trial.

  • By

  • Andrea Surnit

  • June 18, 2026

  • 4 min

Share

Clinical Report: AI Tool May Cut Macular Edema Referrals

Overview

An AI-based optical coherence tomography (AI-OCT) system significantly reduced false-positive referral rates for diabetic macular edema (DME) compared to standard fundus photograph-based screening. The study demonstrated a reduction in unnecessary referrals while maintaining high sensitivity for DME detection.

Background

Diabetic macular edema is a leading cause of vision loss among patients with diabetes, and accurate screening is crucial for timely intervention. Traditional fundus photography often results in high false-positive referral rates, burdening specialist clinics. The introduction of AI technologies in screening may enhance diagnostic accuracy and reduce unnecessary referrals.

Data Highlights

GroupFalse-Positive Referral RateReferral RateReferral Specificity
AI-OCT24%39%87%
Standard Care69%100%0%

Key Findings

  • The AI-OCT group had a false-positive referral rate of 24%, compared to 69% in the standard care group.
  • Referral rates decreased from 100% in the standard care group to 39% in the AI-OCT group.
  • Referral sensitivity was 100% in both groups.
  • Referral specificity was 87% in the AI-OCT group, compared to 0% in the standard care group.
  • The AI-OCT system achieved 99% sensitivity and 91% specificity for DME detection in a validation study.
  • All patients in the standard care group were referred for specialist evaluation due to the existing referral pathway.

Clinical Implications

The AI-OCT system may serve as an effective adjunct to traditional screening methods, potentially reducing the number of unnecessary referrals for DME evaluation. Clinicians should consider integrating AI technologies into their screening protocols to enhance diagnostic accuracy.

Conclusion

The use of an AI-OCT system in diabetic macular edema screening shows promise in reducing false-positive referrals while maintaining high sensitivity. Further studies are needed to assess real-world implementation and clinical outcomes.

Related Resources & Content

  1. JAMA, 2023 -- An AI-Based OCT System to Detect Diabetic Macular Edema: A Prospective Validation and Noninferiority Randomized Clinical Trial
  2. HKMJ, 2024 -- Are we making good use of our public resources? The false-positive rate of screening by fundus photography for diabetic macular oedema
  3. Diabetes Care, 2024 -- Performance of Artificial Intelligence in Detecting Diabetic Macular Edema From Fundus Photography and Optical Coherence Tomography Images: A Systematic Review and Meta-analysis
  4. Ophthalmology Management — Detecting an Elusive Cause For Macular Edema
  5. Retinal Physician — Treatment of Uveitic Macular Edema
  6. Retinal Physician — RPS: From the Podium to the Practice
  7. Retinal Physician — Treating Macular Edema
  8. Retinopathy, Neuropathy, and Foot Care: Standards of Care in Diabetes—2026
  9. Diabetic Retinopathy Preferred Practice Pattern® - Ophthalmology
  10. Performance of Artificial Intelligence in Detecting Diabetic Macular Edema From Fundus Photography and Optical Coherence Tomography Images: A Systematic Review and Meta-analysis | Diabetes Care | American Diabetes Association
  11. Are we making good use of our public resources? The false-positive rate of screening by fundus photography for diabetic macular oedema | HKMJ
  12. Clinical setting-dependent diagnostic accuracy of artificial intelligence and store-and-forward diabetic retinopathy screening: a systematic review and meta-analysis | npj Digital Medicine
  13. An AI-Based OCT System to Detect Diabetic Macular Edema: A Prospective Validation and Noninferiority Randomized Clinical Trial | Artificial Intelligence | JAMA | JAMA Network

Original Source(s)

AI Tool May Cut Macular Edema Referrals

  • by Andrea Surnit

  • June 18, 2026

Related Content

ACP Suggests First-Line Obesity Drugs

Semaglutide and tirzepatide had the greatest net benefit among evaluated treatments, the committee concluded.

OI Therapy Raised BMD, Not Fracture Benefit

Teriparatide followed by zoledronic acid increased bone mineral density but did not reduce fracture risk compared with standard care in adults with osteogenesis imperfecta.

Sugar-Sweetened Beverages May Raise Risk of  Hepatic Cancer Subtypes

In a pooled analysis of more than 1.5 million patients, sugar-sweetened beverages were associated with higher risks of HCC and ICC, while artificially-sweetened beverages showed no independent association with hepatic cancer.