NIH R21/R33: Transformative Co-Robotic Technology for Autism Intervention

描述（由申请人提供）：根据疾病控制和预防中心（CDC）最近对ASD儿童患病率的估计，ASD儿童的患病率为1/88，有效的早期识别和治疗通常被描述为公共卫生紧急情况。鉴于目前干预科学的局限性和整个生命周期中疾病的巨大成本，迫切需要更有效的治疗方法来提高干预效果。越来越多的研究调查了先进的交互技术在ASD干预中的应用，包括计算机技术，虚拟现实（VR）环境，以及最近的机器人系统。拟议研究的主要目标是设计和初步测试一个强大的机器人干预平台和环境，专门用于加速核心ASD缺陷早期领域的改善。特别是，它专注于开发一种机器人干预系统（ARIA：自适应机器人介导的干预架构），能够将实时，非侵入性的凝视检测与智能环境无缝集成，该智能环境具有根据儿童表现自主改变系统功能的能力，以影响早期联合注意力技能，被认为是ASD病因学和治疗的基本社会沟通构建模块。虽然技术进步不太可能取代传统的干预模式，但我们假设自适应机器人系统可能具有巨大的价值，因为它是潜在的加速技术，可以以有效的方式增强学习干预模式，特别是对于那些对非生物行为和事件的应急反应表现出强大差异的儿童，这些行为和事件相对于在很小的时候与社会伙伴的互动。在拟议的研究中，我们将通过明确的设计和测试这样一个系统，以提高早期联合注意力技能领域内的性能，研究机器人技术对ASD幼儿的现实潜力。因此，我们项目的R21阶段的具体目标和里程碑集中在：1）实现非接触式凝视技术集成到系统中，2）实现我们的系统架构的自主闭环操作，以及3）在与自主系统交互时展示足够的初始用户功能。在实现这些里程碑并展示初始系统能力之后，我们项目的R33阶段将在试点干预实验期间评估机器人介导架构的性能。本试验将专门评价与ASD幼儿相关的系统能力，以及与方案中招募和保留相关的可行性数据、用户耐受性和患者满意度。这一试点试验还将评估系统内联合注意力技能的改善，以及评估在更大规模试验中推广这种技能的可能性的方法。

DESCRIPTION (provided by applicant): With the most recent Centers for Disease Control and Prevention (CDC) prevalence estimates for children with ASD at 1 in 88, effective early identification and treatment is often characterized as a public health emergency. Given the present limits of intervention science and the enormous costs of the disorder across the lifespan, there is an urgent need for more efficacious treatments that can enhance intervention outcomes. A growing number of studies have investigated the application of advanced interactive technologies to ASD intervention, including computer technology, virtual reality (VR) environments, and more recently, robotic systems. The primary goal of the proposed research is the design and preliminary testing of a robust robotic intervention platform and environment specifically designed to accelerate improvements in early areas of core ASD deficit. In particular, it focuses on developing a robotic intervention system (ARIA: Adaptive Robot-mediated Intervention Architecture) capable of seamlessly integrating real-time, non-invasive detection of gaze with an intelligent environment endowed with the ability to autonomously alter system function based on child performance to impact early joint attention skills, thought to be fundamental social communication building blocks central to etiology and treatment of ASD. While it is unlikely that technological advances will take the place of traditional intervention paradigms, we hypothesize that adaptive robotic systems may hold great value as potential accelerant technologies that enhance learning intervention modalities in potent ways, particularly for children who show powerful differences in their contingent responses to non-biological actions and events relative to interactions with social partners at very early ages. In the proposed research, we will investigate the realistic potential of robotic technology for young children with ASD via explicit design and tests of such a system to improve performance within the domain of early joint attention skills. Thus, the specific aims and milestones of the R21 phase of our project focus on: 1) Achieving integration of noncontact gaze technology into the system, 2) realizing autonomous closed-loop operation of our system architecture, and 3) demonstrating adequate initial user functioning while interacting with the autonomous system. Subsequent to attaining these milestones and demonstrating initial system capacity, the R33 phase of our project will assess the performance of the robot-mediated architecture during a pilot intervention experiment. This trial will specifically evaluate system capacities as related t very young children with ASD as well as feasibility data related to recruitment and retention in the protocol, user tolerance, and patient satisfaction. This pilot trial will also assess within system improvement in joint attention skills, as well as a methodology for assessing potential generalization of such skills in a larger trial.