Collaborative Research: Deconstructing the contributions of muscle intrinsic mechanics to control of locomotion using a novel Muscle Avatar approach

运动动物在复杂的环境中表现出令人印象深刻的耐力、速度和敏捷性。与人类设计的机器相比，动物的运动尤其令人印象深刻。与动物相比，目前机器人、假肢和外骨骼的稳定性、灵活性和能源经济性仍然很差。动物性能与技术之间的显著差距，部分源于对肌肉生理学和机械功能理解的根本差距。肌肉是动物肌肉骨骼系统中唯一能动控制的组织，因此在使能和控制运动中起着中心作用。然而，过去20年的发展使人们越来越认识到，肌肉生理学和运动科学的重要问题仍然没有解决，该领域的理论基础仍然不完整。特别是，在动态和扰动运动条件下建模和预测肌肉功能的能力仍然很差。该项目将结合创新的实验技术和建模方法，开发新的肌肉模型，可以解释和预测在广泛条件下的肌肉运动。这些发现有可能改变许多领域——使神经科学家、生物学家、临床医生和生物医学工程师能够提出有关人类和动物行为、运动控制、肌肉和骨骼功能以及神经系统和肌肉变化能力的问题。研究小组将与临床和工程领域的同事合作，将研究结果应用于人类康复、疾病和损伤的治疗，以及假肢和外骨骼设备等辅助技术的设计和控制。在动物神经力学领域，“自上而下”的方法（关注动物整体行为，但缺乏对潜在机制的洞察）与“自下而上”的方法（描述机制，但缺乏对动物行为贡献的洞察）之间存在着明显的差距。该团队将开发新的工具来弥补这一差距：1)包括固有粘弹性特性的预测肌肉模型；2)跨层次整合的实验方法。该项目新颖的“肌肉化身”方法将有助于弥合这一差距，并能够对内在肌肉特性和神经激活对运动控制的贡献进行严格分析。目的1测试肌肉化身方法在小鼠和珍珠鸡肌肉中复制稳定和扰动的体内工作回路模式的能力。在目标2中，在稳定和受干扰的情况下，将测量珍珠鸡肌肉在体内的肌肉劳损、激活和力，肌肉虚拟形象将用于定量评估内在肌肉特性和神经驱动如何有助于稳定反应。在Aim 3中，将开发替代肌肉模型，并比较每种模型在高应变和扰动收缩条件下预测体内肌肉功能的能力。计算可处理的肌肉模型对于运动的闭环神经力学模拟至关重要，它越来越多地用于了解肌肉功能和感觉运动控制如何响应衰老，损伤和神经肌肉疾病而变化。研究结果可以为临床康复策略和辅助装置的设计提供信息。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Moving animals achieve impressive athletic feats of endurance, speed, and agility in complex environments. Animal locomotion is particularly impressive in contrast to that of human-engineered machines. The stability, agility and energy economy of current robots, prostheses and exoskeletons remains poor compared to animals. This pronounced gap between animal performance and technology stems, in part, from fundamental gaps in the understanding of muscle physiology and mechanical function. Muscle is the only actively controlled tissue in animal musculoskeletal systems, and therefore plays a central role in enabling and controlling movement. Yet, developments over the past 20 years have led to growing recognition that important problems in muscle physiology and movement sciences remain unsolved and the theoretical foundation of the field remains incomplete. In particular, the ability to model and predict muscle function under dynamic and perturbed locomotor conditions remains poor. This project will combine innovative experimental techniques with modeling approaches to develop new muscle models that can explain and predict muscle movement under a broad range of conditions. The findings have potential to transform numerous fields— enabling neuroscientists, biologists, clinicians and biomedical engineers to ask questions about human and animal behavior, control of motion, function of muscles and bones, and capacity of the nervous system and muscles to change. The research team will collaborate with colleagues in clinical and engineering fields to translate the findings into applications in human rehabilitation, treatment of disease and injury, and the design and control of assistive technology such as prosthetics and exoskeleton devices.In the field of animal neuromechanics, a pronounced gap exists between ‘top down’ approaches — those that focus on whole-animal behavior but lack insight into underlying mechanisms— vs ‘bottom-up’ approaches — those that characterize mechanisms but lack insight into their contributions to animal behavior. The team will develop new tools to bridge this gap: 1) predictive muscle models that include intrinsic viscoelastic properties; and 2) experimental approaches that integrate across levels. This project’s novel ‘muscle avatar’ approach will help bridge this gap, and enable rigorous analysis of intrinsic muscle property and neural activation contributions to control of locomotion. Aim 1 tests the ability of the muscle avatar approach to replicate steady and perturbed in vivo work loop patterns in mouse and guinea fowl muscles. In Aim 2, in vivo muscle strain, activation, and force will be measured during steady and perturbed running in guinea fowl muscles, and the muscle avatar will be used to quantitatively assess how intrinsic muscle properties and neural drive each contribute to stabilizing responses. In Aim 3, alternative muscle models will be developed, and the ability of each model to predict in vivo muscle function in high strain and perturbed contraction conditions will be compared. Computationally tractable muscle models are essential for closed loop neuromechanical simulations of locomotion, which are increasingly used to understand how muscle function and sensorimotor control change in response to aging, injury and neuromuscular disorders. The findings could inform clinical rehabilitation strategies and the design of assistive devices.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.