SCH: Multimodal,Task-Aware Movement Assessment and Control: Clinic to Home

SCH：多模式、任务感知运动评估和控制：诊所到家庭

• 批准号: 10019455
• 负责人: Louis N Awad
• 金额: $ 29.02万
• 依托单位: BOSTON UNIVERSITY (CHARLES RIVER CAMPUS)
• 依托单位国家: 美国
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-30 至 2023-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10019455
• 关键词: Address; Adult; Algorithms; Awareness; Caregivers; Clinic; Clinical; Communities; Data; Detection; Development; Devices; Diagnosis; Distant; Emerging Technologies; Environment; Evaluation; Event; Foundations; Frail Elderly; Functional disorder; Goals; Healthcare Systems; Home environment; Hospitalization; Hospitals; Impairment; Independent Living; Individual; Institution; Instruction; Intervention; Laboratories; Learning Module; Life; Lower Extremity; Machine Learning; Measurement; Modality; Modeling; Monitor; Motion; Movement; Muscle; Patients; Physical activity; Physiological; Population; Protocols documentation; Robot; Robotics; Self-Help Devices; Standardization; System; Techniques; Technology; Testing; Time; Training; Vision; Walking; Work; accurate diagnosis; aging in place; base; cost; frailty; functional decline; functional electrical stimulation; functional status; human model; loss of function; multimodality; neuroprosthesis; next generation; novel; prevent; recruit; response; sensor; technology development; tool; wearable sensor technology

We propose to develop a novel, distributed sensor platform that continuously assesses movement in the background of one's life with the goal of helping people age in place and avoid expensive and lengthy hospitalizations. On the one hand, the platform will combine measurements from a heterogeneous and complementary set of inertial, physiological , and vision sensors with state-of-the-art techniques from robotics and machine learning, together with clinically informed dynamic models of human motion. On the other hand, the platform will use these data to target the prompt detection of the mobility deficits that often precipitate the onset of frailty, with the goal of facilitating personalized caregiver alerts if a decline in functional status is detected. Moreover, the platform will provide context-aware control inputs to facilitate unconstrained use of powered assistive technologies in the home. This project has three main thrusts: assessment, control, and home intervention. In the assessment component, our work will extend well-proven techniques of multi-modal sensor fusion for mapping and localization of robots to home-based movement monitoring and intervention. The novelty of this work lies in the tight integration of machine learning modules for real-time activity recognition and movement dysfunction diagnosis. In the control component, our work will push the boundaries of what is possible with current powered assistive devices by developing novel control mechanisms that take advantage of the new capabilities provided by the estimation component (e.g., adapting control to changes in activities and environmental contexts). In the home intervention component, we will collect data that will refine the sensing and control algorithms and involve caregivers in alerts. A patient-in-the-loop development approach will be utilized where domain-informed protocols will generate the data necessary to train and evaluate our system, both in the clinic and in the home. By enabling timely detection of movement dysfunction and facilitating unconstrained use of powered assistive technologies, this foundational technology has paradigm-disrupting potential to prevent the onset of frailty and alter the treatment options for frail individuals. In parallel, the estimation component of the system could be used in clinical settings to automate and standardize time-intensive and highly subjective functional movement assessments, allowing more accurate diagnoses while freeing clinicians for other important tasks. RELEVANCE (See instructions): Frail older adults constitute the sickest, most expensive, and fastest growing segment of the US population. Home-based technologies that facilitate aging in place and reduce high-cost, hospital- and institution-based interventions are desperately needed. Our proposed distributed sensor platform has the potential to address this need by enabling the timely detection of the mobility deficits that often precipitate the onset of frailty and proactive caregiver and technological interventions that can delay, or prevent, mobility loss.

我们建议开发一种新颖的分布式传感器平台，该平台可以连续评估物体中的运动 一个人的生活背景，目标是帮助人们原地变老，避免昂贵和漫长的时间 住院治疗。一方面，该平台将结合来自异构和 一组互补的惯性、生理和视觉传感器，采用机器人技术的最先进技术 和机器学习，以及临床知情的人体运动动态模型。另一方面， 该平台将使用这些数据来及时检测经常导致行动不便的流动性缺陷。 虚弱发作，目的是在功能状态下降时向护理人员发出个性化警报 检测到。此外，该平台将提供上下文感知控制输入，以方便不受限制地使用 家庭动力辅助技术。 该项目有三个主要重点：评估、控制和家庭干预。在评估中 组件，我们的工作将扩展经过充分验证的多模态传感器融合技术，用于测绘和 将机器人本地化到家庭运动监控和干预。这部作品的新颖之处在于 机器学习模块的紧密集成，用于实时活动识别和运动功能障碍 诊断。在控制组件中，我们的工作将突破当前的可能范围 通过开发利用新技术的新型控制机制来驱动辅助设备 估计组件提供的能力（例如，使控制适应活动的变化和 环境背景）。在家庭干预部分，我们将收集数据来完善传感 和控制算法并使护理人员参与警报。患者循环开发方法将是 用于领域知情协议将生成训练和评估我们的系统所需的数据， 无论是在诊所还是在家里。 通过及时检测运动功能障碍并促进不受限制地使用动力辅助设备 技术，这种基础技术具有颠覆范式的潜力，可以防止衰弱的发生 并改变体弱者的治疗方案。同时，系统的估计组件可以 用于临床环境中，以自动化和标准化时间密集型和高度主观的功能 运动评估，可以实现更准确的诊断，同时让临床医生能够执行其他重要任务。 相关性（参见说明）： 体弱的老年人是美国人口中病情最严重、花费最高且增长最快的群体。 基于家庭的技术，可促进就地养老并降低医院和机构的高成本 迫切需要干预措施。我们提出的分布式传感器平台有潜力 通过及时发现通常会导致发生的流动性缺陷来满足这一需求 虚弱和主动的护理人员以及可以延迟或预防行动能力丧失的技术干预措施。