基于人工智能的眼底照相和光相干断层扫描双模态分析在社区医院糖尿病视网膜病变筛查中的应用。

BACKGROUND: To assess the feasibility and clinical utility of artificial intelligence (AI)-based screening for diabetic retinopathy (DR) and macular edema (ME) by combining fundus photos and optical coherence tomography (OCT) images in a community hospital. METHODS: Fundus photos and OCT images were taken for 600 diabetic patients in a community hospital. Ophthalmologists graded these fundus photos according to the International Clinical Diabetic Retinopathy (ICDR) Severity Scale as the ground truth. Two existing trained AI models were used to automatically classify the fundus images into DR grades according to ICDR, and to detect concomitant ME from OCT images, respectively. The criteria for referral were DR grades 2-4 and/or the presence of ME. The sensitivity and specificity of AI grading were evaluated. The number of referable DR cases confirmed by ophthalmologists and AI was calculated, respectively. RESULTS: DR was detected in 81 (13.5%) participants by ophthalmologists and in 94 (15.6%) by AI, and 45 (7.5%) and 53 (8.8%) participants were diagnosed with referable DR by ophthalmologists and by AI, respectively. The sensitivity, specificity and area under the curve (AUC) of AI for detecting DR were 91.67%, 96.92% and 0.944, respectively. For detecting referable DR, the sensitivity, specificity and AUC of AI were 97.78%, 98.38% and 0.981, respectively. ME was detected from OCT images in 49 (8.2%) participants by ophthalmologists and in 57 (9.5%) by AI, and the sensitivity, specificity and AUC of AI were 91.30%, 97.46% and 0.944, respectively. When combining fundus photos and OCT images, the number of referrals identified by ophthalmologists increased from 45 to 75 and from 53 to 85 by AI. CONCLUSION: AI-based DR screening has high sensitivity and specificity and may feasibly improve the referral rate of community DR.

背景：在社区医院中，通过结合眼底照片和光相干断层扫描（OCT）图像，评估基于人工智能（AI）的糖尿病视网膜病变（DR）和黄斑水肿（ME）筛查的可行性和临床实用性。

方法：对社区医院的 600 名糖尿病患者进行眼底照相和 OCT 检查。眼科医生根据国际临床糖尿病视网膜病变（ICDR）严重程度分级标准对这些眼底照片进行分级，作为金标准。使用两种现有的训练有素的 AI 模型，根据 ICDR 自动将眼底图像分类为 DR 分级，并分别从 OCT 图像中检测并发 ME。转诊标准为 DR 分级 2-4 级和/或存在 ME。评估 AI 分级的敏感性和特异性。分别计算眼科医生和 AI 确定的可转诊 DR 病例数。

结果：眼科医生在 81 名（13.5%）参与者中发现 DR，在 94 名（15.6%）参与者中发现 AI，45 名（7.5%）和 53 名（8.8%）参与者被眼科医生和 AI 诊断为可转诊 DR。AI 检测 DR 的敏感性、特异性和曲线下面积（AUC）分别为 91.67%、96.92%和 0.944。对于检测可转诊 DR，AI 的敏感性、特异性和 AUC 分别为 97.78%、98.38%和 0.981。眼科医生从 49 名（8.2%）参与者的 OCT 图像中发现 ME，AI 从 57 名（9.5%）参与者的 OCT 图像中发现 ME，AI 的敏感性、特异性和 AUC 分别为 91.30%、97.46%和 0.944。当结合眼底照片和 OCT 图像时，眼科医生确定的转诊人数从 45 人增加到 75 人，AI 从 53 人增加到 85 人。

结论：基于 AI 的 DR 筛查具有较高的敏感性和特异性，可能可行地提高社区 DR 的转诊率。