RetinoScan

DiabeatIt

RetinoScan

A portable fundus imaging device with integrated machine learning for automated Diabetic Retinopathy detection.

Noida, Uttar Pradesh, India

IND

Nonprofit

Diabetic Retinopathy (DR) is a vision-threatening complication of diabetes that damages the blood vessels of the retina. According to IDF, over 537 million adults were living with diabetes in 2021. Approximately one in five people with diabetes develops DR, and one in ten suffers from vision-threatening stages. A major challenge lies in the lack of accessible screening and diagnostic services, especially in low-resource settings. Globally, over 75% of people with diabetes live in low- and middle-income countries, where healthcare infrastructure is often strained. In India, for instance, there are only 11 skilled eye specialists available for every million people. In states such as Uttar Pradesh, most primary healthcare centres lack ophthalmic equipment, and the shortage of retinal specialists delays diagnosis and treatment. DR often progresses silently, with patients unaware until vision loss is irreversible. WHO advocates for the integration of DR screening into routine diabetes care, yet many countries fall short of implementing this. Without early detection and timely care, DR will continue to be a leading cause of preventable blindness—especially in underserved populations where the burden is the most.

Our solution, RetinoScan, is a compact, AI-powered device designed for real-time detection and classification of Diabetic Retinopathy (DR). It integrates a custom optical imaging system with a deep learning model to enable accessible and efficient retinal screening. To ensure portability and usability in low-resource settings, RetinoScan is integrated into a compact optical setup using aspheric, achromatic doublet, plano-convex, and biconvex lenses for high-resolution retinal imaging.A built-in camera module within the apparatus captures the retinal images. These images are analyzed using a Swin-L Transformer-based AI model, trained to classify DR into five stages: No DR, Mild, Moderate, Severe, and Proliferative DR. The model leverages self-attention mechanisms and hierarchical feature learning to achieve high accuracy. It processes the images locally and instantly displays results through a user-friendly interface.The hardware and AI work together to provide instant, automated diagnosis without needing a specialist, bridging gaps in accessibility and expertise. RetinoScan is designed for deployment in primary healthcare centers, mobile camps, and teleophthalmology networks, making DR screening scalable, cost-effective, and accessible. By combining innovative optics and advanced AI, our solution aligns with global goals to prevent avoidable blindness through early detection and timely intervention.

Our initial target is 250,000 diabetic individuals in rural and semi-urban regions of Uttar Pradesh, India’s most populous state. The next phase will extend our outreach to high-burden states such as Bihar, Madhya Pradesh, and Rajasthan—impacting over 20 million people at risk of Diabetic Retinopathy (DR). Ultimately, our vision is to reach 2 million diabetic individuals across India’s underserved regions, especially those reliant on government-run primary health centers. A striking example of the care gap comes from rural Uttar Pradesh, where diabetic patients are often referred to distant urban hospitals for retinal screening, facing delays of weeks or months due to limited specialist availability. Many are lost to follow-up, resulting in missed diagnoses and avoidable blindness. This is the critical window where RetinoScan intervenes. By equipping local health workers with our portable, AI-powered device, DR screening can be performed instantly at the point of care—with results available within minutes. Our solution dramatically increases screening rates, enables timely referrals, and reduces reliance on specialists. Over the next three years, we aim to screen 2 million individuals in rural India. RetinoScan stands to transform diabetic eye care by making early detection affordable, scalable, and accessible where it is needed most.