Merge of motion analysis, multibody dynamics and finite element method for the subject-specific analysis of cartilage loading patterns during gait: differences between rotation and moment-driven models of human knee joint

Understanding joint loading is important when evaluating sports training methods, sports equipment design, preventive training regimens, post-op recovery procedures, or in osteoarthritis' etiology research. A number of methods have been introduced to estimate joint loads but they have been limited by the lack of accuracy in the joint models, including primarily the lack of patient-specific motion inputs in the models with sophisticated, fibril-reinforced material models. The method reported here records and applies patient-specific human motion for in-depth cartilage stress estimation. First, the motion analysis of a subject was conducted. Due to skin motion, multibody simulation was used to correct motion capture. These data was used as an input in a finite element model. The model geometry was based on magnetic resonance imaging and cartilage was modeled as a fibril-reinforced poroviscoelastic material. Based on the experimental motion data (motion analysis and multibody simulation), two models were created: a rotation-controlled and a moment-controlled model. For comparison, a model with motion input from the literature was created. The rotation-controlled model showed the most even stress distribution between lateral and medial compartments and smallest stresses and strains in a depth-wise manner. The model based on the literature motion simulated very high stresses and uneven stress distribution between the joint compartments. Our new approach to determine dynamic knee cartilage loading enables estimations of stresses and strains for a specific subject over the entire motion cycle.