Researchers have determined that artificial intelligence (AI)-assisted non-mydriatic point-of-care screening administered during primary care visits may help patients with diabetes adhere to follow-up eye care. They also believe an automated diabetic retinopathy screening system can help reduce referrals for patients with low-risk features.
The study included 180 patients with diabetes, all of whom underwent fundus photography, followed by automated retinal image analysis with human supervision. The system referred patients with positive or inconclusive screening results for a comprehensive ophthalmic evaluation. With the automated screening, 8.3% had referable diabetic eye disease, 13.3% had vision-threatening disease and 29.4% had an inconclusive result. The remaining 48.9% had negative screening results, confirmed by human readers, and were not referred for follow-up ophthalmic evaluation.
Overall, the automated system showed a sensitivity of 100% in detecting an abnormal screening result, while its specificity was 65.67%. Among patients referred for a one-year follow-up ophthalmic evaluation, the researchers found that adherence rate was 55.4%, compared with the historical adherence rate of 18.7%.
While many believe retinal screening examinations to be costly, the researchers believe this option can benefit primary care clinics that serve low-income metropolitan patient populations.
Liu J, Gibson E, Ramchal S, et al. Diabetic retinopathy screening with automated retinal image analysis in a primary care setting improves adherence to ophthalmic care. Opthalmol Retina. June 17, 2020. [Epub ahead of print].