卢旺达基于人工智能的糖尿病视网膜病变筛查的可行性和可接受性。

Feasibility and acceptance of artificial intelligence-based diabetic retinopathy screening in Rwanda.

机构信息

Clinical Services, Orbis International, New York, New York, USA.

RIIO iHospital, Rwanda International Institute of Ophthalmology, Kigali, Rwanda.

出版信息

Br J Ophthalmol. 2024 May 21;108(6):840-845. doi: 10.1136/bjo-2022-322683.

DOI:10.1136/bjo-2022-322683

PMID:37541766

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.

摘要

背景

需要了解人工智能 (AI) 辅助的糖尿病视网膜病变 (DR) 筛查的实际应用情况。

方法

2021 年 3 月至 6 月，在卢旺达的 4 家糖尿病诊所，使用 AI 解读的视网膜成像对同意参与的患者进行 DR 筛查。此外，图像由经过英国国民保健系统认证的视网膜图像分级器进行分级。基于国际糖尿病视网膜病变分类的 DR 分级为 2.0 或更高被认为是有转诊必要的。该 AI 系统旨在检测 DR 以外的视神经和黄斑异常。AI 还将垂直杯盘比为 0.7 及以上和/或在 60%及以上的截断值识别到的黄斑异常视为有转诊必要。

结果

在通过 AI 筛查的 827 名参与者（59.6%为女性（n=493））中，33.2%（n=275）被转诊进行后续检查。对 AI 筛查的满意度很高（99.5%，n=823），63.7%的参与者（n=527）更喜欢 AI 而不是人工分级。与人工分级相比，AI 对有转诊必要的 DR 的敏感度为 92%（95%CI 0.863%，0.968%），特异性为 85%（95%CI 0.751%，0.882%）。AI 转诊的患者中：88 例（32.0%）为单纯 DR，109 例（39.6%）为 DR 合并异常，65 例（23.6%）为单纯异常，13 例（4.73%）为其他原因。DR 转诊患者的就诊率最高，为 53.4%。