Russian AI Platform Accelerates Early Detection of Diabetic Retinopathy

The new AI‑powered system analyzes retinal scans with over 97% accuracy, enabling ophthalmologists to spot diabetic eye disease sooner and reduce the risk of blindness.

Russian researchers at Skoltech and the University of Sharjah have unveiled an artificial‑intelligence platform that automates the analysis of retinal images to diagnose diabetic retinopathy. Driven by the need for rapid, reliable screening, the system processes high‑resolution scans and identifies pathological changes in the retina and blood vessels — tasks that previously required painstaking manual segmentation.

Diabetic retinopathy, a complication of diabetes, can silently progress to complete vision loss if not caught early. The newly developed neural network examines the retina with remarkable depth and precision, achieving diagnostic accuracy in excess of 97%. By automating vessel and microvascular assessment, the platform eliminates the need for clinicians to manually trace tiny capillaries, slashing analysis time and minimizing human error.

Small blood vessels often exhibit the first signs of retinal disease, making microvascular detection critical. The AI’s enhanced sensitivity in recognizing these subtle changes not only speeds up diagnosis of diabetic retinopathy, but also opens the door to earlier intervention for other ocular pathologies. Faster, more accurate assessments pave the way for timely treatment plans and smoother rehabilitation processes.

With its imminent deployment in clinical settings, this AI solution promises to transform eye care by boosting both the efficiency and effectiveness of retinal diagnostics. Early, precise evaluation of the retina can lead to immediate treatment, reduce the likelihood of severe complications, and help preserve patients’ vision.