Autonomous AI to mitigate disparities for diabetic retinopathy screening in youth during and after COVID-19

项目摘要 2018年，FDA批准了第一款自主医疗人工智能(AI-HC)软件，该软件可以解释 用非散瞳眼底相机拍摄的视网膜图像，为糖尿病视网膜病变提供即时结果 (DR)在护理点(POC)对成年糖尿病患者进行筛查。家长奖的PI是第一个 在儿科实施这项技术，展示安全性、有效性和公平性，并节省成本 病人。他们还发现，少数族裔青年、家庭收入较低和医疗补助保险的人 接受推荐筛查的可能性较小，但更有可能获得父母奖 假设在糖尿病护理环境中实施POC自主人工智能将增加DR筛查 减少青年糖尿病患者的发病率，减少在获得筛查方面的差距，并具有成本效益地获得医疗保健 系统。在父母奖中，AIM 1是一项在两个临床地点进行的随机对照试验，以确定是否自主 与眼部护理专业人员(ECP)相比，AI增加了筛查，如果AI筛查呈阳性的人 更有可能去ECP寻求后续行动。AIM 2是一项人工智能筛查的前瞻性观察性试验 确定人工智能是否缩小了筛查中的差异，并提高了高危、少数和低风险人群的比例 收入，年轻人谁去跟进，如果他们的人工智能屏幕是积极的。如果人工智能被证明提高了筛查率，而 减少获得护理方面的差距，它有可能重塑筛查方法。 与家长奖合作提供了一个独特的机会来解决两个紧迫的伦理问题： 为人工智能进行临床试验有哪些伦理挑战？您如何预测、识别和 在AI-HC造成伤害之前解决它们的道德问题？补充剂团队与 家长在F.D.A.审查期间奖励研究人员，在眼科合作社区 为FDA提供成像，并从这些经验中开发一种方法来识别和解决 人工智能在医疗保健方面的伦理挑战。在本增刊中，我们将试用我们的伦理分析方法：1) 当AI-HC用于DR的临床试验正在进行时，实时确定出现的伦理问题 通过家长奖进行；2)就如何解决这些已确定的道德问题达成专家共识 家长奖面临的挑战；以及3)在执行1和2中，改进可概括的方法来确定和 通过AI-HC应对伦理挑战，以及如何解决未来伦理问题的路线图 AI-HC的临床试验。

Project Summary In 2018 the FDA approved the first autonomous healthcare artificial intelligence (AI-HC) software that interprets retinal images taken with a non-mydriatic fundus camera, providing an immediate result for diabetic retinopathy (DR) screening at the point of care (POC) for adults with diabetes. The PIs of the parent award were the first to implement this technology in pediatrics, demonstrating safety, effectiveness and equity, and cost-savings to the patient. They also found that minority youth, those with lower household income and Medicaid insurance were less likely to undergo recommended screening, yet were more likely to have DR. The parent award hypothesizes that implementing POC autonomous AI in the diabetes care setting will increase DR screening rates in youth with diabetes, mitigate disparities in access to screening, and be cost-effective to the health care system. In the parent award, Aim 1 is a randomized control trial at two clinic sites to determine if autonomous AI increases screening compared to an eye-care professional (ECP), and if those who screen positive by AI are more likely to go for follow-up at the ECP. Aim 2 is a prospective observational trial of AI screening to determine if AI mitigates disparities in screening, and improves the proportion of at-risk, minority and low income, youth who go for follow-up if their AI screen is positive. If AI is shown to increase screening rates while mitigating disparities in access to care, it has the potential to reshape screening methods. Partnering with the parent award presents a unique opportunity to address two pressing ethical questions: What are the ethical challenges with conducting clinical trials for AI? How do you anticipate, identify, and address ethical problems with AI-HC before they cause harm? The supplement team has worked closely with the parent award investigators during their F.D.A. review, in the Collaborative Community on Ophthalmic Imaging for the F.D.A. and, from these experiences, in development of a methodology to identify and address ethical challenges with AI for healthcare. In this supplement we will pilot our ethical analysis methodology to: 1) identify the ethical issues emerging with clinical trials of AI-HC for DR in real time as such trials are being conducted through the parent award; 2) develop expert consensus on how to address these identified ethical challenges for the parent award; and 3) in doing 1 & 2, refine a generalizable approach for identifying and addressing ethical challenges with an AI-HC and a roadmap for how to address ethical concerns with future clinical trials of AI-HC.