基于人工智能图像分析、反射性散瞳与图像复核的糖尿病视网膜病变远程医疗筛查结局
单机构回顾性分析1052例在初级保健机构接受糖尿病视网膜病变(DR)筛查的连续成人患者,采用IDx-DR人工智能系统分析免散瞳眼底照片,对不可分级图像行1%托吡卡胺反射性散瞳后重复AI分析,并对所有图像进行人工复核。965例(91.7%)图像可被AI分级,与人工复核判定需进一步散瞳检查的超过轻度非增殖性DR相比,AI分级敏感度100%、特异度89.2%、阳性预测值27.5%、阴性预测值100%;图像可分级率随年龄增长而降低。
要点
- 965例中AI判读阳性138例(14.3%),阴性827例(85.7%),与人工复核相比敏感度100%、阴性预测值100%。
- 免散瞳图像可分级者580例(55.1%),免散瞳不可分级的472例中440例(93.2%)接受反射性散瞳。
- 图像可分级率与年龄负相关:<70岁者93.5%可分级(61.9%免散瞳),≥70岁者85.3%可分级(31.0%免散瞳),P < .001。
核心结论
- 在结合免散瞳眼底照相与按需药物散瞳的方案中,将基于人工智能的图像分析纳入真实世界初级保健DR筛查未出现假阴性结果,并提供良好的图像可分级性。
- 图像可分级性随患者年龄增长而降低。
研究设计与样本
- 效度与信度分析,单机构回顾性审查18个月内连续接受DR照片筛查的1052例成人患者。
- 采用IDx-DR人工智能系统分析免散瞳眼底照片,免散瞳图像不可分级时行1%托吡卡胺反射性散瞳并重复AI分析,所有图像均行人工复核。
主要发现
- 1052例中965例(91.7%)图像可被AI分级,其中580例(55.1%)免散瞳图像可分级,免散瞳不可分级的472例中440例(93.2%)接受反射性散瞳。
- 965例中AI判读阳性(超过轻度NPDR)138例(14.3%),阴性827例(85.7%)。
- 与人工复核判定需进一步散瞳检查的超过轻度NPDR相比,AI分级敏感度100%(95%CI 90.8-100%)、特异度89.2%(95%CI 87.0-91.1%)、阳性预测值27.5%(95%CI 24.0-31.4%)、阴性预测值100%(95%CI 99.6-100%)。
- 图像可分级率与年龄负相关:<70岁者93.5%可分级(61.9%免散瞳),≥70岁者85.3%可分级(31.0%免散瞳),P < .001。
关键数值与推荐
- AI判读阳性定义为超过轻度NPDR,人工复核参照标准为需进一步散瞳检查的超过轻度NPDR。
- 免散瞳图像不可分级时采用1%托吡卡胺行反射性散瞳后重复AI分析。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
PURPOSE: To examine real-world telemedicine outcomes of diabetic retinopathy (DR) screening with artificial intelligence (AI)-based image analysis, reflex dilation, and secondary image overread in a primary care setting. DESIGN: Validity and reliability analysis. METHODS: Single institution review of 1052 consecutive adult patients who received diabetic retinopathy photoscreening in the primary care setting over an 18-month period. Nonmydriatic fundus photographs were acquired and analyzed by the IDx-DR AI-based system. When nonmydriatic images were ungradable, reflex dilation (1% tropicamide) and mydriatic photography were performed for repeat AI-based analysis. Manual overread was performed on all images. Patient demographics, clinical characteristics, and screening outcomes were recorded. RESULTS: A total of 965 of 1052 patients (91.7%) had AI-gradable fundus photographs: 580 had gradable nonmydriatic imaging (55.1%) and 440 of 472 patients with ungradable nonmydriatic photographs had reflex dilation (93.2%). One hundred thirty-eight of 965 patients (14.3%) were AI-graded as "positive" (greater than mild NPDR) and 827 of 965 were "negative" (85.7%), with 100% sensitivity (95% CI 90.8-100%), 89.2% specificity (95% CI 87.0-91.1%), 27.5% positive predictive value (95% CI 24.0-31.4%), and 100% negative predictive value (95% CI 99.6-100%) compared with manual overread assessment of greater than mild NPDR requiring further evaluation with a comprehensive dilated examination. Image gradeability was inversely related to patient age: 93.5% gradable (61.9% nonmydriatic) for patients aged <70 years vs 85.3% (31.0% nonmydriatic) for patients aged 70+ years (P < .001). CONCLUSION: Incorporation of AI-based image analysis into real-world primary care diabetic retinopathy screening yielded no false negative results and offered excellent image gradeability within a protocol combining nonmydriatic fundus photography and pharmacologic dilation, as needed. Image gradeability was lower with increasing patient age.
引用
- Mehra AA, Softing A, Guner MK, et al. Diabetic Retinopathy Telemedicine Outcomes With Artificial Intelligence-Based Image Analysis, Reflex Dilation, and Image Overread. Am J Ophthalmol. 2022. PMID: 35970206. doi:10.1016/j.ajo.2022.08.008
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。