Feasibility and acceptance of artificial intelligence-based diabetic retinopathy screening in Rwanda.
Whitestone Noelle, Nkurikiye John, Patnaik Jennifer L, Jaccard Nicolas, Lanouette Gabriella, Cherwek David H, Congdon Nathan, Mathenge Wanjiku
AI Summary
AI-based DR screening in Rwanda was accurate for referable DR (92% sensitivity) and highly accepted, demonstrating its practical utility for identifying patients needing follow-up, including for other eye anomalies.
Abstract
Background
Evidence on the practical application of artificial intelligence (AI)-based diabetic retinopathy (DR) screening is needed.
Methods
Consented participants were screened for DR using retinal imaging with AI interpretation from March 2021 to June 2021 at four diabetes clinics in Rwanda. Additionally, images were graded by a UK National Health System-certified retinal image grader. DR grades based on the International Classification of Diabetic Retinopathy with a grade of 2.0 or higher were considered referable. The AI system was designed to detect optic nerve and macular anomalies outside of DR. A vertical cup to disc ratio of 0.7 and higher and/or macular anomalies recognised at a cut-off of 60% and higher were also considered referable by AI.
Results
Among 827 participants (59.6% women (n=493)) screened by AI, 33.2% (n=275) were referred for follow-up. Satisfaction with AI screening was high (99.5%, n=823), and 63.7% of participants (n=527) preferred AI over human grading. Compared with human grading, the sensitivity of the AI for referable DR was 92% (95% CI 0.863%, 0.968%), with a specificity of 85% (95% CI 0.751%, 0.882%). Of the participants referred by AI: 88 (32.0%) were for DR only, 109 (39.6%) for DR and an anomaly, 65 (23.6%) for an anomaly only and 13 (4.73%) for other reasons. Adherence to referrals was highest for those referred for DR at 53.4%.
Conclusion
DR screening using AI led to accurate referrals from diabetes clinics in Rwanda and high rates of participant satisfaction, suggesting AI screening for DR is practical and acceptable.
MeSH Terms
Shields Classification
Key Concepts6
Diabetic retinopathy (DR) screening using artificial intelligence (AI) in diabetes clinics in Rwanda led to accurate referrals and high rates of participant satisfaction, suggesting AI screening for DR is practical and acceptable.
Among 827 participants screened for diabetic retinopathy (DR) by artificial intelligence (AI) in diabetes clinics in Rwanda, 33.2% (n=275) were referred for follow-up.
Satisfaction with artificial intelligence (AI) screening for diabetic retinopathy (DR) was high (99.5%, n=823) among participants in diabetes clinics in Rwanda, and 63.7% of participants (n=527) preferred AI over human grading.
Compared with human grading, the sensitivity of the artificial intelligence (AI) system for referable diabetic retinopathy (DR) was 92% (95% CI 0.863%, 0.968%), with a specificity of 85% (95% CI 0.751%, 0.882%) in a screening program in Rwanda.
Of the participants referred by artificial intelligence (AI) for diabetic retinopathy (DR) screening in Rwanda, 88 (32.0%) were for DR only, 109 (39.6%) for DR and an anomaly, 65 (23.6%) for an anomaly only, and 13 (4.73%) for other reasons.
Adherence to referrals for diabetic retinopathy (DR) screening using artificial intelligence (AI) in Rwanda was highest for those referred for DR at 53.4%.
Related Articles5
Comparison of Deep Learning and Clinician Performance for Detecting Referable Glaucoma from Fundus Photographs in a Safety Net Population.
Cohort StudyEvaluating a Foundation Artificial Intelligence Model for Glaucoma Detection Using Color Fundus Photographs.
Observational StudyTime to Glaucoma Progression Detection by Optical Coherence Tomography in Individuals of African and European Descents.
Cohort StudyA generalised computer vision model for improved glaucoma screening using fundus images.
Observational StudyOptical Coherence Tomography Versus Optic Disc Photo Assessment in Glaucoma Screening.
ReviewIs this article assigned to the wrong chapter(s)? Let us know.