AI system shows high accuracy for diabetic retinopathy screening
"AI could help reduce the burden on ophthalmology services by triaging large numbers of patients with diabetes and allowing specialists to focus on those who most urgently need care."
To evaluate the accuracy of an AI-based system for identifying referable diabetic retinopathy in a clinical setting, with implications for improving patient care.
Key Findings:
The AI system maintained very high diagnostic accuracy, exceeding regulatory benchmarks for sensitivity and specificity, which were not specified.
Strong agreement was found between AI and human grading, with the AI correctly identifying all cases of vision-threatening diabetic retinopathy.
23.5% of patients were referred for newly detected ocular abnormalities, with only 15.7% due to diabetic retinopathy.
Interpretation:
The findings suggest that AI can effectively triage diabetic retinopathy screenings, potentially alleviating the burden on ophthalmology services and improving patient access to care, ultimately enhancing patient outcomes.
Limitations:
Further research is needed to assess long-term outcomes, cost-effectiveness, and the need for diverse clinical settings to enhance generalizability.
The study was conducted in a single clinical setting, which may limit generalizability.
Conclusion:
Integrating AI into routine diabetes care could enhance screening accessibility, help prevent avoidable blindness, and reduce the burden on healthcare systems.
AACE Endocrine AI
is published by Conexiant under a license arrangement with the American Association of Clinical Endocrinology, Inc. (AACE®). The ideas and opinions expressed in
AACE Endocrine AI do not necessarily reflect those of Conexiant or AACE. For more information, see Policies.