期刊文章%@ 1438-8871 %I JMIR出版物%V 24 %N卡塔尔世界杯8强波胆分析 9 %P e40249% T医务人员和住院医生在眼病筛查中使用深度学习的偏好:离散选择实验%阿林，森林%阿李，黎平%阿邹，海东%阿徐，伊%阿陆，莉娜%+上海市眼病防治中心，上海市虹桥路1440号，86 02162539696,lulina781019@qq.com %K离散选择实验%K偏好%K人工智能%K AI %K视力健康%K筛查%D 2022 %7 20.9.2022 %9原创论文%J J医学互联网Res %G英文%X背景:深度学习辅助眼病诊断技术在眼病筛查中的应用越来越广泛。然而，目前还没有研究表明医疗服务提供者和居民愿意使用它的先决条件。目的:本文旨在揭示医疗服务提供者和居民对使用人工智能(AI)进行社区眼病筛查的偏好，特别是对准确性的偏好。方法:采用离散选择实验方法，对上海市卫生保健机构和居民进行问卷调查。共有34家具有充分人工智能辅助筛查经验的医疗机构参与。共有39名医务人员和318名住院医师回答问卷，并对不同属性的筛查策略进行权衡，包括漏诊率、超诊率、筛查结果反馈效率、眼科医生参与程度、组织形式、成本、筛查结果反馈形式。采用条件logit模型和逐步选择方法估计偏好。结果:医护人员首选准确性较高的深度学习模型，其特异性应大于90%(10%超诊断的比值比[OR]=0.61; P<.001), which was much higher than the Food and Drug Administration standards. However, accuracy was not the residents’ preference. Rather, they preferred to have the doctors involved in the screening process. In addition, when compared with a fully manual diagnosis, AI technology was more favored by the medical staff (OR=2.08 for semiautomated AI model and OR=2.39 for fully automated AI model; P<.001), while the residents were in disfavor of the AI technology without doctors’ supervision (OR=0.24; P<.001). Conclusions: Deep learning model under doctors’ supervision is strongly recommended, and the specificity of the model should be more than 90%. In addition, digital transformation should help medical staff move away from heavy and repetitive work and spend more time on communicating with residents. %M 36125854 %R 10.2196/40249 %U //www.mybigtv.com/2022/9/e40249 %U https://doi.org/10.2196/40249 %U http://www.ncbi.nlm.nih.gov/pubmed/36125854