Exploiting reciprocity between constraints and instantaneous motion to reconstruct individual knee kinematics

Musculoskeletal models are limited in their clinical application by the complexity of their personalization. In this work we present a novel and general approach for the reconstruction of the individual joint kinematics that may provide a reference for the personalization of more complex models. The approach relies on the principle of virtual work and obtains the joint motion by imposing reciprocity between instantaneous helical axis and joint constraints. As a case of study, we investigate a cadaveric knee, whose motion was experimentally measured for validation. Comparison between computed and experimental motion resulted in maximum mean rotational and translational errors of 2.6 ± 0.9° and 1.2 ± 0.7 mm, respectively, supporting the efficacy of the presented approach.