Simulating Muscle Spasticity in Children with Cerebral Palsy

We developed three models of spasticity based on feedback from muscle states. The first model relied on feedback from muscle length and velocity. The second model relied on feedback from muscle length, velocity and acceleration. The third model relied on feedback from muscle force and its first time derivative (force rate). We first calibrated the models based on experimental data collected during passive stretches in children with cerebral palsy. We then used the calibrated models to predict the spastic response during gait. We showed that only the model based on feedback from muscle force and force rate could explain spastic muscle activity during passive stretches and gait. This suggests that force encoding in muscle spindles in combination with altered feedback gains and thresholds underlie activity of spastic muscles during passive stretches and gait.This project contains experimental data and code necessary to reproduce all results presented in the associated publication. Please find more information in the manual (see folder Manual in simcpspasticity-code). <br/><br/>This project includes the following software/data packages: <br/> <ul> <li> <a href="https://simtk.org/frs?group_id=1499#pack_2061">Project code </a> : This project contains code for simulating muscle spasticity during passive stretches and gait in children with cerebral palsy. </li> <li> <a href="https://simtk.org/frs?group_id=1499#pack_2062">Project data </a> : This package contains experimental data for simulating muscle spasticity during passive stretches and gait in children with cerebral palsy. The package is divided into 12 folders containing data for each subject separately. You should re-organize the data as described in the manual (see project code). The data include:1) EMGs, GRFs, joint angles (obtained with Inverse Kinematics), joint torques (obtained with Inverse Dynamics) and muscle analyses from gait trials. 2) EMGs, joint angles (obtained with Inverse Kinematics), joint torques (obtained with Inverse Dynamics) and muscle analyses from passive stretches of hamstrings and gastrocnemii at different velocities (IPSA trials). 3) Results from muscle-tendon parameter estimation based on EMG. 4) Results from spasticity model calibration based on passive stretches. 5) Results from spastic response prediction during gait. </li> </ul>

本研究团队基于肌肉状态反馈，构建了三种痉挛模型。首种模型依赖肌肉长度与速度的反馈；次种模型则基于肌肉长度、速度及加速度的反馈；第三种模型则依赖于肌肉力量及其一阶导数（力量率）的反馈。我们首先基于儿童脑瘫患者被动拉伸过程中的实验数据对模型进行校准。随后，我们运用校准后的模型预测步行过程中的痉挛反应。研究表明，唯有基于肌肉力量与力量率反馈的模型能够解释被动拉伸及步行过程中的痉挛肌肉活动。此发现暗示，肌肉梭中的力量编码与反馈增益及阈值的改变共同构成了被动拉伸及步行过程中痉挛肌肉活动的基础。本项目包含用于重现相关出版物中所有结果的实验数据与代码。更详细的信息请参阅手册（详见simcpspasticity-code文件夹中的Manual）。

本项目包含以下软件/数据包：

<ul>
<li><a href="https://simtk.org/frs?group_id=1499#pack_2061">项目代码</a>：本项目包含模拟脑瘫儿童被动拉伸及步行过程中肌肉痉挛的代码。</li>
<li><a href="https://simtk.org/frs?group_id=1499#pack_2062">项目数据</a>：此数据包包含模拟脑瘫儿童被动拉伸及步行过程中肌肉痉挛的实验数据。数据包分为12个文件夹，分别包含每个受试者的数据。您应按照手册中的说明（详见项目代码）重新组织数据。数据包括：1) 电磁肌电图（EMGs）、地面反作用力（GRFs）、关节角度（通过逆运动学获得）、关节扭矩（通过逆动力学获得）及步行试验中的肌肉分析。2) 电磁肌电图、关节角度（通过逆运动学获得）、关节扭矩（通过逆动力学获得）及不同速度下腘绳肌和比目鱼肌的被动拉伸肌肉分析。3) 基于电磁肌电图的肌肉-肌腱参数估计结果。4) 基于被动拉伸的痉挛模型校准结果。5) 步行过程中的痉挛反应预测结果。</li>
</ul>