VIRTUAL MUSCLE: A HIERARCHICAL MATHEMATICAL MUSCLE MODEL

拟议中的SBIR项目的目的是让C-Motion在交互程序(虚拟肌肉)中增强强大的数学肌肉模型(由曾傑瑞·勒布博士的实验室开发)并将其商业化，以试验神经肌肉控制的动态模拟。它将允许研究人员使用图形用户界面和自动化测试协议，在正向动态模拟中对可定制肌肉模型的行为进行经验探索。在第一阶段，我们将：确定人体肌肉的模型参数；在Visual C++中实施模型，优化计算速度，并在商业建模程序中作为执行器实施；开发模拟单个肌肉收缩的测试程序，并测试输出对模型参数扰动的敏感度：开发定制的人体和猴子手臂运动模拟，Loeb博士及其同事目前正在研究；设计自动化测试的命令脚本计划；以及定义测试数据后期处理和编译成格式化报告的规范。在第二阶段，我们将开发一个项目，探索计算作用于关节的单个肌肉力量的方法，从而导致对运动障碍的临床和研究评估。建议的商业应用：一种易于使用并准确捕捉真实肌肉和肌腱的复杂机械特性的商业肌肉模型，将对必须了解神经肌肉控制以解决临床康复问题的研究人员有用。虚拟肌肉将成为下一代临床运动分析软件的关键元素，为神经肌肉控制的研究提供环境，以开发解决一般分布问题和估计个人肌肉力量的新策略。

The purpose of the proposed SBIR project is for C-Motion to enhance and commercialize a powerful mathematical muscle model (developed in the laboratory of Dr. Jerry Loeb) within an interactive program (Virtual Muscle) for experimenting with dynamic simulations of neuromuscular control. It will allow investigators to explore empirically the behavior of customizable muscle models in forward dynamic simulations, using a graphical user interface and automated testing protocols. In Phase I, we will: determine model parameters for human muscles; implement the model in Visual C++, optimizing for speed of the calculations and allowing implementation as an actuator in commercial modeling programs; develop a testing program for simulation of the contraction of a single muscle, and test the output sensitivity to perturbations in model parameters: develop a custom simulation of two- dimensional human and monkey arm movements now being studied by Dr. Loeb and colleagues; design a command script scheme for automated testing; and define the specifications for post-processing and compilation of test data into a formatted report. In Phase II we will develop a program for exploring approaches to calculating individual muscle forces acting at a joint, leading to clinical and research evaluations of movement disorders. PROPOSED COMMERCIAL APPLICATIONS: A commercial muscle model that is easy to use, and captures accurately the complex mechanical properties of real muscles and tendons, will be useful to researchers who must understand neuromuscular control in order to address clinical rehabilitation issues. Virtual Muscle will be a key element of the next generation of clinical movement analysis software providing an environment for research in neuromuscular control in the development of new strategies for solving the general distribution problem and estimating individual muscle forces.