Achieving Optimal Motor Function in Stroke Survivors via a Human-Centered Approach to Design an mHealth Platform

项目总结/摘要 上肢轻瘫是中风后最常见的损伤，影响75%的中风幸存者， 其在两个分支中的一个中可能更突出。大多数功能障碍的恢复发生在 中风后的头几周以及之后的稳定期。不幸的是，即使在患者达到稳定期后， 恢复期，患肢功能水平可能下降。因此，在临床上， 通过继续以下方式，在自发恢复的前几周后维持恢复的功能水平： 在日常生活中练习使用患肢。 可穿戴技术已经成为一种低成本、客观的工具，用于监测鞋面的性能。 日常生活活动（ADL）。然而，到目前为止，还没有研究调查 移动健康（mHealth）系统的有效性，旨在实现慢性高剂量运动性能 中风幸存者在现实世界中的情况。具体来说，就是目标设定的最优配置，反馈 在利益相关者（患者和临床医生）之间共享数据的机制和方式仍然未知。 该提案旨在开发和验证移动医疗技术，旨在鼓励受影响的肢体 在慢性卒中幸存者进行ADL期间使用。为了实现这一目标，我们将采用 独特的指戴式环形传感器（加速度计），由我们的学术界和工业界合作开发，可以 捕获在ADL表现中必不可少的肢体的粗臂和细手使用。我们将 研究做出积极行为改变的重要方面，以鼓励充分使用受影响的肢体， 利用从传感器数据得出的计算见解与来自提供者的临床见解相结合。 为了实现这一研究目标，Aim 1将专注于开发一个移动健康平台，包括 身体联网传感器和基于云的系统，以监测慢性中风患者肢体的实际使用情况 幸存者在目标2中，我们将开发基于机器学习的算法，以提取临床上有意义的 来自传感器数据的关于真实世界上肢使用的信息。目标3将研究的优化设计， 我们的移动健康系统-如个人定制的目标，设计的反馈，媒体和时间， 提供反馈，以及在利益相关者（患者和临床医生）之间共享数据的方法-通过以人为本 设计方法最后，在目标4中，我们将验证移动健康的短期（8周）有效性 通过现场部署研究，改进系统对受影响肢体的使用。 我们相信，该项目的成果将打开一扇新的大门，通往以前未探索的数据集， 了解患者-技术相互作用，以促进积极的行为改变，从而实现高剂量 物理和职业治疗，这可以形成广泛的未来调查的基础， 偏瘫康复和个性化疾病管理。

Project Summary / Abstract Upper-limb paresis is the most common impairment following a stroke affecting 75% of stroke survivors, which can be more prominent in one of the two limbs. Most recovery of functional impairments occurs within the first few weeks after stroke and plateaus thereafter. Unfortunately, even after patients reach a stable phase of recovery, their functional level of the stroke-affected limb may decline. Therefore, it is clinically important to maintain the regained functional level beyond the first couple weeks of spontaneous recovery by continuing to practice the use of the affected limb during daily living. Wearable technologies have emerged as a low-cost, objective tool to monitor the performance of the upper limbs during activities of daily living (ADLs). However, to date, there exists no study that has investigated the effectiveness of a mobile-health (mHealth) system aiming to enable high-dosage motor performance in chronic stroke survivors in the real-world setting. Specifically, the optimal configuration of the goal setting, feedback mechanism and ways to share data among the stakeholders (patients and clinicians) remains unknown. This proposal aims to develop and validate an mHealth technology that aims to encourage affected limb use during the performance of ADLs in chronic stroke survivors. To accomplish this goal, we will employ the unique finger-worn ring sensor (accelerometer), developed by our academic-industry partnership, that can capture both gross-arm and fine-hand use of the limbs that are essential in the performance of ADLs. We will study important aspects of making positive behavior changes to encourage the affected limb use by fully leveraging the computational insights drawn from sensor data combined with clinical insights from providers. To accomplish this research goal, Aim 1 will focus on the development of an mHealth platform, composed of body-networked sensors and cloud-based systems, to monitor the real-world use of the limbs in chronic stroke survivors. In Aim 2, we will develop machine-learning based algorithms to extract clinically meaningful information regarding real-world upper limb use from sensor data. Aim 3 will investigate the optimal design of our mHealth system – such as individual tailoring of the goal, design of the feedback, medium and timing to deliver feedback, and ways to share data among the stakeholders (patients and clinicians) – via human-centered design approaches. Finally, in Aim 4, we will validate the short-term (8 weeks) effectiveness of the mHealth system in improving the use of the affected limb through a field deployment study. We believe that outcomes of this project will open a new door leading to previously unexplored datasets and understanding of patient-technology interactions to promote positive behavior changes to enable a high dosage of physical and occupational therapy, which can form the basis of a wide range of future investigations of hemiparesis rehabilitation and personalized disease management.