Tools for the Biomechanical Analysis of Human Movement

人体运动生物力学分析工具

• 批准号: RGPIN-2021-03095
• 负责人: Deluzio, Kevin
• 金额: $ 3.35万
• 依托单位: Queen's University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2021
• 资助国家: 加拿大
• 起止时间: 2021-01-01 至 2022-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=742377
• 关键词: Tools; Biomechanical; Analysis; Human; Movement

My research career is focused on improving the utility of human movement data by creating innovative engineering tools for its analysis, classification and interpretation. Working with my students, and supported by Discovery Grant funding, my research has contributed significantly to the development of classification tools to detect differences in movement patterns and computational models to give these differences physical meaning. In addition to improving applied science, these tools have provided insight into the fundamental processes underlying a given movement pattern. The influence of this work is evident by the uptake of these methods in human movement research; however, there remains large barriers to the integration of human movement data in applied research and practical applications. Identifying and resolving these barriers is what currently motivates me and where the target of my current and future research has been set. Markerless motion capture is an emerging technology for estimating the 3D position and orientation (pose) of a human multibody model. Data can now be instantly collected anywhere that video cameras can record the subject, while virtually eliminating preparation time for the subject. There is no contact between the subject and the researcher, and subjects' data are collected while wearing their own clothing, removing the barrier of requiring minimal tight fitting clothing for research. To fully realize the potential of markerless motion capture and to take advantage of the new flexibility in raw data, the focus of my research program will be developing the next generation of biomechanical analysis tools. Increasing the automation of human movement data collection as a result of the adoption of computer vision and machine learning will substantially increase our ability to generate high quality data at volume in a fraction of the time for a fraction of the cost. Large scale, longitudinal data collections that have never before been possible in biomechanics due to the many limitations of existing technology will become achievable. We are developing tools that can assess movement throughout the natural environment. Eliminating unnecessary physical proximity through a completely contactless approach to biomechanical data collection is a significant advantage and will help researchers recruit a more diverse group of subjects because it overcomes many barriers to participation, including eliminating the need for minimalistic tight-fitting clothing, and the travel required to get to a motion laboratory. This work may finally allow the adoption of biomechanical tools into wider use, improving the study of clinical movement disorders, musculoskeletal diseases, injuries, and athletic performance.

我的研究生涯专注于通过创建用于分析、分类和解释的创新工程工具来提高人体运动数据的实用性。我的研究与我的学生一起工作，在发现基金的支持下，为检测运动模式差异的分类工具和赋予这些差异物理意义的计算模型的开发做出了重大贡献。除了改进应用科学之外，这些工具还提供了对给定运动模式背后的基本过程的深入了解。这项工作的影响通过这些方法在人体运动研究中的采用而显而易见。然而，将人体运动数据整合到应用研究和实际应用中仍然存在很大障碍。识别并解决这些障碍是我当前的动力，也是我当前和未来研究的目标。无标记动作捕捉是一种新兴技术，用于估计人体多体模型的 3D 位置和方向（姿势）。现在可以在摄像机可以记录主题的任何地方立即收集数据，同时几乎消除了主题的准备时间。受试者和研究人员之间没有接触，受试者的数据是在穿着自己的衣服时收集的，消除了研究时需要最少紧身衣服的障碍。为了充分发挥无标记动作捕捉的潜力并利用原始数据的新灵活性，我的研究计划的重点将是开发下一代生物力学分析工具。 由于采用计算机视觉和机器学习，提高了人体运动数据收集的自动化程度，这将大大提高我们在很短的时间内以很少的成本生成大量高质量数据的能力。由于现有技术的许多限制，生物力学中以前从未实现过的大规模纵向数据收集将成为可能。我们正在开发可以评估整个自然环境中运动的工具。通过完全非接触式的生物力学数据收集方法消除不必要的物理接近是一个显着的优势，并将帮助研究人员招募更多样化的受试者群体，因为它克服了参与的许多障碍，包括消除对简约紧身衣服的需要，以及前往运动实验室所需的旅行。这项工作最终可能使生物力学工具得到更广泛的应用，改善临床运动障碍、肌肉骨骼疾病、损伤和运动表现的研究。