A digital health approach to early identification and outcome monitoring in autism

摘要-项目1 杜克大学自闭症卓越中心的总体目标是使用翻译的数字健康和计算 解决对更有效的自闭症筛查工具、客观结果衡量标准、 以及基于大脑的生物标记物，可用于自闭症儿童的临床试验。项目1的目标 是评估基于计算机视觉分析和机器学习的新型数字行为评估工具 这可以在现实世界中实施，以提高自闭症筛查的准确性并实现可扩展、 目的对儿童行为发育进行纵向监测。项目1将招募大量人口 通过初级保健诊所对16至30个月大的蹒跚学步的儿童进行评估 用于检测自闭症早期症状的新型数字表型应用程序(APP)。该应用程序会自动量化 使用计算机视觉分析直接观察儿童的行为，并部署在智能手机或 平板电脑。我们将评估敏感度、特异度、阴性/阳性预测值和重测可靠性 当父母在家交付时，用于自闭症检测的数字表型应用程序。我们亦会评估 APP对16-30、36和48个月的自闭症儿童的纵向结局监测的可用性 与标准化临床测量方法比较，检验其收敛效度。目标是扩大 可用于自闭症筛查和结果监测的行为测量类型，我们将探索 使用计算机视觉分析从录制的视频中测量亲子互动的可行性 回家。最后，我们将与项目2合作，设计一个自动化的初级临床决策支持系统 护理提供者将自闭症筛查信息与有关转诊的可行指导整合在一起 诊断和服务，并评估其感知的可用性。我们的长期愿景是改变临床护理的方式 是通过提供创新的解决方案来提供的，这些解决方案解决了长期存在的获得护理的障碍。

ABSTRACT – Project 1 The overall goal of the Duke Autism Center of Excellence is to use a translational digital health and computational approach to address the critical need for more effective autism screening tools, objective outcome measures, and brain-based biomarkers that can be used in clinical trials with young autistic children. The goal of Project 1 is to evaluate novel digital behavioral assessment tools based on computer vision analysis and machine learning that can be implemented in real-world settings to improve the accuracy of autism screening and enable scalable, objective longitudinal monitoring of children’s behavior and development. Project 1 will recruit a large population of 16- to 30-month-old toddlers through primary care clinics to evaluate the accuracy of a remotely administered novel digital phenotyping application (app) for detecting early signs of autism. The app automatically quantifies direct observations of children’s behavior using computer vision analysis and is deployed on a smartphone or tablet. We will assess the sensitivity, specificity, negative/positive predictive values, and test-retest reliability of the digital phenotyping app for autism detection when delivered by parents at home. We will also assess the app’s usability for longitudinal outcome monitoring of autistic children at 16-30, 36, and 48 months of age by examining its convergent validity compared to standardized clinical measures. With a goal of expanding the types of behavioral measures that could be used for autism screening and outcome monitoring, we will explore the feasibility of using computer vision analysis to measure parent-child interaction from videos recorded at home. Finally, in collaboration with Project 2, we will design an automated clinical decision support for primary care providers that integrates autism screening information with actionable guidance regarding referrals for diagnosis and services and assess its perceived usability. Our long-term vision is to transform how clinical care is delivered by providing innovative solutions that address long-standing barriers in access to care.