Estimation of subject-specific ligament length variation during knee flexion

Abstract

Detailed knowledge about subject-specific knee ligament length variation during knee flexion is crucial for musculoskeletal modeling in human movement analysis and in clinical contexts (Ascani et al. 2014). In this respect, accurate estimation of ligament origin and insertion locations is needed. Although extensive literature exists describing the ligament attachment site position ex vivo (Kopf et al. 2009), this information is hardly obtainable in vivo unless using medical imaging techniques (magnetic resonance, MRI, or computed tomography, CT) that are time- consuming, expensive and, when radiation is involved, risky for the subjects. Furthermore, the identification of the ligament attachment site location using MRI may be characterized by significant inter- and intra-observer variability thus leading to errors that are too large to generate reliable subject-specific knee models (Rachmat et al. 2014). Therefore, the aim of the present study is to propose and validate a method for the accurate in vivo estimation of the subject-specific length of the major knee ligaments (anterior and posterior cruciate, ACL and PCL, medial and collateral ligaments, MCL and LCL) which does not necessarily require the use of medical imaging techniques to identify the ligament attachment sites.