TrueKneeMov

The knee pathologies are common, multiple, and very disabling. Quantifying the pre, per and post operative motor function is essential for a good care of the patients. If per operative navigation has become a standard, the pre or post operative motion capture, yet based on the same technologies, is not. Navigation directly tracks the skeleton while motion capture tracks skin markers. This introduces a measurement bias (called soft tissue artefact) which has never been properly accounted for. True knee motion remains impracticable. The goal of this project is to explore the most promising compensation methods (bias identification, introduction of joint constraints). This is based on (a) the acquisition of a unique database coupling bi-plane fluoroscopy and motion capture in different tasks (e.g., stepping) and for different populations (pathological or not) and this is based on (b) original biomechanical models of the skin deformation and of the movements and/or efforts in the joints. A multi-objective optimisation under uncertainty will be performed. It is hypothesised that the true tri-dimensional knee motion is a robust non-dominated solution of a two-objective problem: tracking the skin markers while maintaining the joint integrity.