Locomotor Apparatus Bioengineering Laboratory, Department of Human Movement and Sport Sciences, University of Rome Foro Italico, piazza Lauro De Bosis 15, 00135 Rome, Italy.
Gait Posture. 2011 Apr;33(4):706-11. doi: 10.1016/j.gaitpost.2011.03.006. Epub 2011 Apr 1.
When using skin markers and stereophotogrammetry for movement analysis, bone pose estimation may be performed using multi-body optimization with the intent of reducing the effect of soft tissue artefacts. When the joint of interest is the knee, improvement of this approach requires defining subject-specific relevant kinematic constraints. The aim of this work was to provide these constraints in the form of plausible values for the distances between origin and insertion of the main ligaments (ligament lengths), during loaded healthy knee flexion, taking into account the indeterminacies associated with landmark identification during anatomical calibration. Ligament attachment sites were identified through virtual palpation on digital bone templates. Attachments sites were estimated for six knee specimens by matching the femur and tibia templates to low-dose stereoradiography images. Movement data were obtained using stereophotogrammetry and pin markers. Relevant ligament lengths for the anterior and posterior cruciate, lateral collateral, and deep and superficial bundles of the medial collateral ligaments (ACL, PCL, LCL, MCLdeep, MCLsup) were calculated. The effect of landmark identification variability was evaluated performing a Monte Carlo simulation on the coordinates of the origin-insertion centroids. The ACL and LCL lengths were found to decrease, and the MCLdeep length to increase significantly during flexion, while variations in PCL and MCLsup length was concealed by the experimental indeterminacy. An analytical model is given that provides subject-specific plausible ligament length variations as functions of the knee flexion angle and that can be incorporated in a multi-body optimization procedure.
当使用皮肤标记和体视摄影进行运动分析时，可以使用多体优化来进行骨骼姿势估计，目的是减少软组织伪影的影响。当感兴趣的关节是膝关节时，这种方法的改进需要定义特定于主题的相关运动学约束。这项工作的目的是在考虑到在解剖校准过程中标志点识别的不确定性的情况下，为主要韧带（韧带长度）的起点和止点之间的距离提供合理的约束值，这些约束值是以加载健康膝关节屈曲时的合理值的形式给出的。通过在数字骨骼模板上进行虚拟触诊来确定韧带附着点。通过将股骨和胫骨模板与低剂量体视摄影图像匹配，为六个膝关节标本估计了附着点。使用体视摄影和销钉标记获得运动数据。计算了前交叉韧带、后交叉韧带、外侧副韧带和内侧副韧带深层和浅层束（ACL、PCL、LCL、MCLdeep、MCLsup）的相关韧带长度。通过对起源-插入质心的坐标进行蒙特卡罗模拟来评估标志点识别变异性的影响。发现 ACL 和 LCL 的长度在屈曲时减小，而 MCLdeep 的长度显著增加，而 PCL 和 MCLsup 的长度变化则被实验不确定性所掩盖。提供了一个解析模型，该模型将特定于主题的合理韧带长度变化作为膝关节屈曲角度的函数，并且可以纳入多体优化过程。
