Comparison and validation of five scapulothoracic models for correcting soft tissue artefact through multibody optimisation
Résumé
Soft tissue artefact (STA) is one of the most important limitations when studying human movement through marker-based motion capture techniques. Indeed, the relative movement between the reflective skin markers and the underlying bones introduce errors in bone kinematics. Upper limb kinematics is particularly affected by STA, especially when studying scapula. As a result, there is a need for correction in order to use, with confidence, this technique in clinical environments. Different compensation or correction methods exist, such as multibody optimisation (MBO). Using MBO, the studied limb is represented as a kinematic chain composed of rigid segments linked by mechanical joints. The position and orientation of each segment are then optimised under kinematic and rigid body constraints in order to minimise the sum of the squared distances between measured and model-determined markers positions. It has been shown that the correction obtained depends on the joint model and that anatomical- based models are better suited to reach physiological results. The aim of the study was thus to define and validate different upper limb joint models used in a MBO framework focusing on the definition of an anatomical-based scapulothoracic joint.