SCH: kHealth: Semantic Multisensory Mobile Approach to Personalized Asthma Care

项目总结(见说明)： 智能手机和传感器的激增使持续监测生理、环境和公共健康通知成为可能。然而，只有在患者范围内处理大量多感官和多模式观察数据，并使其具有可操作性，以便能够早期发现；更好地诊断、有效预防和治疗疾病；并提高生活质量，个性化数字健康和患者赋权才能成为现实。这项研究将在哮喘的背景下进行，哮喘每年影响3亿人，并夺走全球超过25万人的生命。根据美国疾病控制与预防中心的数据，9.3%的美国儿童患有哮喘。理解哮喘及其治疗是具有挑战性的；它是一种具有主观原因和症状的多因素疾病，现有的诊断和治疗指南可以使用循证行动建议来改进。随着移动和互联传感器的出现，有关哮喘诱因的及时信息已经成为可能，利用这些信息需要准确地理解和量化传感数据在预防和主动评估哮喘方面的作用。在这项由计算机科学家和执业临床医生领导的应用程序中，我们解决了从使用kHealth技术监控的数据中获取可操作信息的中心问题。通过一项试点研究，我们在临床医生的指导下，在几名儿科患者身上测试了我们配备的kHealth试剂盒，以进行可靠的数据收集和患者操作，取得了良好的初步结果。我们在此基础上，在医学知识的背景下，解决了跨越生理、环境数据和在线数据的异类观测的语义表示、集成、抽象和个性化解释问题。我们的主要假设是，循证治疗哮喘的方法可以帮助医生更准确地确定哮喘的原因、严重程度和控制，并可以提醒患者寻求及时的临床援助，以更好地管理哮喘。

PROJECT SUMMARY (See instructions): The proliferation of smartphones and sensors has made continuous monitoring of physiology, environment, and public health notifications possible. However, personalized digital health and patient empowerment will become reality only if massive multisensory and multimodal observational data is processed within patient context and made actionable to enable early detection; better diagnosis, effective prevention and treatment of the disease; and improve the quality of life. This research will be performed in the context of asthma, which affects 300 million people and claims over 250,000 lives worldwide annually. According to CDC, 9.3% of US children have asthma. Understanding asthma and its management is challenging; it is a multifactorial disease with subjective causes and symptoms, and extant diagnosis and treatment guidelines can be improved using evidence-based action recommendation. With the advent of mobile and connected sensors, timely information about asthma triggers has become available, and its exploitation requires precisely understanding and quantifying the sensed data's role for preventive and proactive assessment of asthma. In this application led by computer scientists and a practicing clinician, we address the central problem of deriving actionable information from data monitored using kHealth technology. Through a pilot study, we have tested our kHealth kit equipped on several pediatric patients under the guidance of a clinical doctor for reliable data collection and patient operation, with promising initial results. We build on this foundation by addressing the problem of semantic representation, integration, abstraction, and personalized interpretation of heterogeneous observations spanning physiological, environmental data, and online data in the context of medical knowledge. Our primary hypothesis is that an evidence-based approach to asthma can help doctors determine more precisely the cause, severity and control of asthma, and can alert patients to seek timely clinical assistance to better manage asthma.