SBIR Phase I: IMPROVING BEHAVIORAL ASSESSMENT OF CHILDREN DIAGNOSED WITH AUTISM SPECTRUM DISORDER THROUGH ENHANCED DATA ACQUISITION

这个SBIR一期项目旨在通过提供目前自闭症治疗提供者无法获得的非临床环境的关键数据来改善被诊断为自闭症谱系障碍（ASD）的儿童的预后。该项目还侧重于加速对这些数据的分析。据估计，在患者的一生中，与ASD相关的费用将超过200万美元。然而，研究表明，早期干预可以将诊断为ASD的儿童的终身护理费用降低2/3。设计一个适当的治疗方案，然后监测其疗效，需要收集一系列数据，以避免依赖推断和轶事。治疗提供者通常在与被诊断为自闭症的个体一起工作时收集详细的数据，但他们指望父母和照顾者向他们提供有关儿童临床环境之外的进展的数据。提交的数据通常是不完整和不充分的（如果提交的话）。这个SBIR第一阶段项目将产生一个原型，它将迅速为自闭症治疗提供者提供关键数据和分析，减少设计最佳治疗方案所需的时间，从而减少与自闭症诊断相关的长期成本。这个第一阶段的项目将产生一个原型，这将使记录数据的过程对父母来说几乎没有摩擦。这些叙事数据将与视频数据、生物特征数据以及从外部被动收集的其他数据相结合。所有的数据将被合并，然后通过机器学习进行分析，以定位模式，并为孩子的行为分析师提供完整的画面。开发一款对自闭症儿童来说不太显眼的相机将是一项挑战，开发相关软件来标记相机视图中的物品并仅保存相关片段（例如崩溃之前，期间和之后的时间）也是一项挑战。开发一种ASD儿童能够忍受的可穿戴设备，以及相关的传感器和软件来跟踪与觉醒和压力相关的指标，也将是一项挑战。然而，最重要的技术挑战将是开发结合多个数据流并输出标准化ABC报告（前事，行为，后果）的软件，该报告将被董事会认证的行为分析师判断为高质量。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。

This SBIR Phase I project intends to improve the outcomes of children diagnosed with autism spectrum disorder (ASD) by providing critical data from non-clinical settings that are currently unavailable to autism treatment providers. The project also focuses on accelerating the analysis of that data. It is estimated that the costs associated with ASD will be more than $2 million over a patient's life. However, studies have shown that the costs of lifetime care for a child diagnosed with ASD can be reduced by 2/3 with early intervention. Designing an appropriate treatment program, and then monitoring its efficacy, requires the collection of a range of data to avoid dependence on inference and anecdotes. Treatment providers often collect detailed data while working with an individual diagnosed with ASD, but they count on parents and caregivers to provide them with data regarding the child's progress outside the clinical setting. The data submitted are usually incomplete and inadequate (if submitted at all). This SBIR Phase I project will result in a prototype that will rapidly provide autism treatment providers with critical data and analysis, reducing the amount of time it takes to design an optimal treatment program, and thereby reducing the long-term costs associated with a diagnosis of ASD.This Phase I project will result in a prototype that will make the process of logging data almost frictionless for parents. That narrative data will be combined with video data, biometric data, and additional data collected passively from external sources. All the data will be merged, and then analyzed via machine learning to locate patterns and provide a complete picture to the child's behavior analyst. Developing a camera that will be unobtrusive enough for a child with ASD to tolerate will present a challenge, as will developing the associated software to tag items within the camera's view and save only relevant clips (such as the times immediately before, during and after a meltdown). It will also be challenging to develop a wearable device that a child with ASD will tolerate, along with the associated sensors and software to track metrics associated with arousal and stress. The most significant technical challenge, however, will be to develop the software that combines the multiple data streams and outputs a standardized ABC report (Antecedent, Behavior, Consequence) that would be judged as high quality by a Board Certified Behavior Analyst.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.