QuBBD: Collaborative Research: Advancing mHealth using Big Data Analytics: Statistical and Dynamical Systems Modeling of Real-Time Adaptive m-Intervention for Pain

随着移动的电话技术的日益普及，实时自适应医疗干预出现了新的机会。 多个“大数据”源的同时增长（例如，移动的健康数据、电子健康记录、实验室测试结果、基因组数据）允许开发个性化推荐。 该奖项支持启动一个合作研究项目，该项目将为慢性病患者的主观疼痛随时间的变化产生一个新的数学模型。该模型将与统计技术相结合，以最终获得优化的，不断更新的治疗计划，平衡疼痛减轻和药物最小化的竞争需求。 这些个性化治疗计划将被纳入目前正在进行的一项关于移动的干预镰状细胞病（SCD）慢性疼痛患者的试点研究。 由于近四分之一的急诊患者就诊的情况本可以作为门诊患者进行管理，因此改善移动的医疗技术以使这些患者能够快速识别和接收适当的医疗保健信息至关重要。目前还没有标准的算法或分析方法来为疼痛等慢性疾病提供实时自适应治疗建议。 此外，当前最先进的方法难以处理使用大数据的连续时间决策优化。 拟议的模型将包括使用微分方程预测未来疼痛水平的动态系统方法，以及将系统参数与患者数据（包括报告的疼痛水平，用药史，个人特征和其他健康记录）联系起来的统计方法。 第三个关键组成部分将是开发和试点研究一种新的控制和优化策略，以平衡减轻疼痛和药物剂量最小化的竞争需求。 该奖项由美国国立卫生研究院大数据到知识（BD2K）计划与国家科学基金会数学科学部合作支持。

With the growing popularity of mobile phone technology, new opportunities have arisen for real-time adaptive medical intervention. The simultaneous growth of multiple "big data" sources (e.g., mobile health data, electronic health records, lab test results, genomic data) allows for the development of personalized recommendations. This award supports initiation of a collaborative research project that will generate a new mathematical model for changes in subjective pain over time in patients with chronic conditions. The model will be combined with statistical techniques to ultimately obtain optimized, continuously-updated treatment plans balancing competing demands of pain reduction and medication minimization. Those resulting personalized treatment plans will be incorporated into a currently active pilot study on mobile intervention in patients living with chronic pain due to sickle cell disease (SCD). Since nearly a quarter of patient visits to the emergency room are for conditions that could have been managed as outpatients, it is crucial to improve mobile health technologies to allow these patients to quickly recognize and receive appropriate health care information. There currently is no standard algorithm or analytical method for real-time adaptive treatment recommendations for chronic conditions like pain. Furthermore, current state-of-the-art methods have difficulty in handling continuous-time decision optimization using big data. The proposed model will consist of a dynamical systems approach using differential equations to forecast future pain levels, as well as a statistical approach tying system parameters to patient data (including reported pain levels, medication history, personal characteristics and other health records). A third key component will be the development and pilot study of a new control and optimization strategy to balance the competing demands of pain reduction and drug dosage minimization. This award is supported by the National Institutes of Health Big Data to Knowledge (BD2K) Initiative in partnership with the National Science Foundation Division of Mathematical Sciences.