USE OF ARTIFICIAL INTELLIGENCE IN SCREENING AND DIAGNOSIS OF DIABETIC RETINOPATHY: REVIEW OF RECENT EVIDENCE

Abstract

Diabetic retinopathy is one of the leading causes of preventable blindness among working-age adults, representing a growing challenge for healthcare systems in light of the global increase in diabetes mellitus prevalence. Early diagnosis is essential to prevent severe visual complications, yet the limited coverage of ophthalmologic examinations in many countries still leads to underdiagnosis and delayed treatment. In this context, artificial intelligence (AI) has emerged as a promising tool for the screening and diagnosis of diabetic retinopathy, providing cost-effective, scalable solutions with sensitivity comparable to that of specialists. Deep learning models have demonstrated the ability to analyze fundus images with speed and accuracy, enabling the implementation of large-scale screening programs, even in regions with a shortage of ophthalmologists. This review compiles recent evidence on the application of AI in diabetic retinopathy screening, highlighting its advantages, limitations, and future perspectives for integration into clinical practice.