Methods to improve the use of wearable sensors in human movement analyses.

步态分析研究对于分析运动表现、病理步态或衰老的影响是重要的。不幸的是，传统的光学步态分析系统是昂贵的，耗时的，并局限于实验室，这限制了他们的可访问性和实际应用。可穿戴传感器为传统系统提供了一种具有成本效益的替代方案，具有在现实条件下收集数据的独特能力。虽然这可能是生物力学研究的新前沿，但可穿戴传感器通常未能实现其在现实世界，不受控制的环境中进行步态分析的潜力。可穿戴传感器的能力与它们目前的用途之间的这种脱节主要是基于处理和管理收集的大量数据的困难。因此，如果可穿戴传感器要发挥其在现实世界中的潜力，就迫切需要解决科学和生物力学界的这些基本方法差距。我的研究计划的长期愿景是为生物力学社区提供新的工具，以帮助使人体运动分析更容易获得。可穿戴传感器为支持这一愿景提供了明确的机会，但迫切需要改进收集、处理、解释和可视化这些数据的方式。因此，本研究的短期目标将集中在i）改进可穿戴传感器数据的活动分类和事件检测算法，ii）开发强大的可穿戴传感器处理管道，iii）评估对不受控制的真实步态模式变化的敏感性，以及iv）开发新的可穿戴传感器数据可视化技术。基于我过去在可穿戴传感器和人体运动分析中的机器学习领域的成功，这项研究将使可穿戴传感器数据在现实世界中的收集，不受控制的设置更高效，更有效，更可解释。因此，这项基础性工作将开发新的分析方法，使研究人员能够收集更好、更有代表性的人体运动数据，从而能够更好地做出有关健康老龄化和肌肉骨骼疾病治疗的明智决策。此外，这项研究将使我的计划发展成为可穿戴技术和人体运动这一新兴领域的世界领导者，同时推动加拿大和我的HQP未来的学术和行业成就。

Gait analysis research is important for analysing sport performance, pathological gait, or the effects of aging. Unfortunately, conventional optical gait analysis systems are expensive, time-consuming, and confined to laboratories, which limits their accessibility and practical application. Wearable sensors offer a cost-effective alternative to conventional systems, with the unique ability to collect data in real-world conditions. While this may be the new frontier of biomechanics research, wearable sensors have generally failed to realize their potential for gait analyses in real-world, uncontrolled settings. This disconnect between what wearable sensors are capable of and what they are currently used for is largely based on the difficulty in processing and managing the large amounts of data collected. Therefore, if wearable sensors are to fulfill their real-world potential, there is a critical need to address these foundational methodological gaps across the scientific and biomechanics community. The long-term vision of my research program is to provide the biomechanics community with new tools to help make human movement analyses more accessible. Wearable sensors provide a clear opportunity to support this vision, but there is an immediate need to improve the ways in which these data are collected, processed, interpreted, and visualized. Therefore, the short-term goals of this research will focus on i) improving activity classification and event detection algorithms for wearable sensor data, ii) developing robust wearable sensor processing pipelines, iii) evaluating the sensitivity to change in uncontrolled, real-world gait patterns, and iv) developing new wearable sensor data visualization techniques. Building on my past successes in the areas of wearable sensors and machine learning in human movement analyses, this research will make the collection of wearable sensor data in real-world, uncontrolled settings more efficient, more effective, and more interpretable. Therefore, this foundational work will develop new analytical methods to empower researchers to collect better and more representative human movement data, which will in turn enable better informed decision-making regarding healthy aging and the treatment of musculoskeletal disorders. Moreover, this research will develop my program as a world-leader in this emerging area of wearable technology and human movement, while driving future academic and industry achievements for Canada and my HQP.