SCC-PG: Bridge: An AI-Enabled Platform to Support Coordinated Care for Children with Autism

患有自闭症谱系障碍（CWA）的儿童通常会出现严重的问题行为，因此需要长期的护理和个性化的临床评估、治疗和干预。应用行为分析（ABA）疗法，被认为是一种基于证据的最佳实践，可能耗时，资源密集，并容易受到人类偏见。这个智能和互联社区规划拨款（SCC-PG）项目旨在共同设计一个基于人工智能的解决方案，用于收集行为数据，以自动测量严重行为（如频率、强度、延迟等），协助诊断和治疗决策，并与患者和家属实时沟通以进行早期干预。该项目团队与一系列社区利益相关者合作，在德克萨斯州圣安东尼奥市进行了一项试点研究，通过人工智能增强平台协调护理CWA。规划过程从确定关键平台参数开始，包括数据类型、隐私问题、主要功能模块和预期性能。研究了不同的人工智能算法，以融合多模态数据来通知ABA。进行了事后调查和焦点小组，以揭示将物联网边缘云原型纳入临床实践的效果。该项目有可能改善自闭症儿童的健康状况及其家庭的福祉。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Children with autism spectrum disorder (CWA) often engage in severe problem behavior, and thus require long-term care with individualized clinical assessment, treatment, and intervention. Applied behavior analysis (ABA) therapy, considered an evidence-based best practice, can be time-consuming, resource intensive, and prone to human bias. This Smart and Connected Communities Planning Grant (SCC-PG) project seeks to co-design an AI-based solution for collecting behavior data for automatic measurement of severe behaviors (e.g., frequency, intensity, latency, etc.), assisting diagnosis and treatment decisions, and communicating to patients and families in real-time for early intervention.The project team partners with a range of community stakeholders in a pilot study in San Antonio, TX for coordinated care of CWA through an AI-augmented platform. The planning process begins with identifying key platform parameters including data types, privacy concerns, main function modules, and expected performance. Different AI algorithms are examined for fusing multi-modal data to inform ABA. Post-survey and focus groups are conducted to uncover the effects of incorporating the IoT-Edge-Cloud prototype into clinical practices. The project has the potential to improve health outcomes for autistic children and the well-being of their families.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.