Remidio’s AI Delivers Strong Real-World Accuracy in Landmark Diabetic Retinopathy Study

The findings show that Remidio’s AI achieved an overall sensitivity of 87 per cent for referable diabetic retinopathy, which includes moderate non-proliferative disease and more advanced stages.

Remidio Innovative Solutions’ diabetic retinopathy AI has demonstrated strong real-world clinical performance in the largest independent global evaluation of its kind, with a Lancet Digital Health study confirming its accuracy, reliability, and suitability for population-scale diabetic eye screening.

The study has analysed more than 200,000 screening encounters and around 1.2 million retinal images, validating the system’s ability to identify sight-threatening disease at scale.

The evaluation has been led by Professor Alicja Rudnicka, Professor Adnan Tufail, and researchers working with the UK National Health Service.

Unlike many earlier AI assessments conducted under controlled research settings, this study has focused on real-world deployment, examining how algorithms perform within routine screening workflows.

Of 32 regulatory-approved diabetic retinopathy AI systems initially reviewed, eight met the criteria for final analysis, including Remidio’s CE Class IIa-approved AI.

The findings show that Remidio’s AI achieved an overall sensitivity of 87 per cent for referable diabetic retinopathy, which includes moderate non-proliferative disease and more advanced stages.

Sensitivity increased substantially for more serious conditions, reaching nearly 99 per cent for moderate-to-severe non-proliferative diabetic retinopathy and about 97 per cent for proliferative disease. These results indicate that the likelihood of missing sight-threatening cases is very low, a key requirement for mass screening programmes.

With a negative predictive value of approximately 98 per cent, the system provides strong reassurance for individuals who screen negative, supporting wider coverage and the possibility of extending screening intervals without increasing clinical risk.

Dr R. Kim, Senior Consultant at Aravind Eye Hospital, said, “Large-scale diabetic retinopathy screening demands both diagnostic sensitivity and operational discipline. The findings reported in this evaluation are clinically meaningful because they demonstrate that AI can detect sight-threatening disease reliably, while maintaining specificity levels that are essential to avoid overwhelming referral systems. This balance is critical for sustainable screening programmes, particularly in high-volume public health settings.”

From a system-level perspective, the study suggests that AI deployment could reduce reliance on manual human grading by up to 80 per cent, allowing ophthalmologists and trained graders to focus on patients who require treatment rather than routine screening.

Dr Divya, Chief Medical Officer at Remidio Innovative Solutions, said, “This study evaluates AI performance at real-world screening scale rather than in controlled environments. The results show that Remidio’s AI can reliably identify sight-threatening disease while avoiding unnecessary referrals, which is critical for safe and efficient public health screening.”

The evaluation also found that Remidio’s AI performance remained consistent across age, sex, and ethnicity, supporting equitable use across diverse populations and addressing concerns about demographic bias in medical AI systems.

Stay tuned for more such updates on Digital Health News