Computational Representation of the Patellofemoral Joint

Many people suffer from knee pain due to abnormal function of the patellofemoral joint and are not able to enjoy normal activities of daily living. Surgical treatments are available and new methods are being developed by the medical industry. However, computational tools to efficiently evaluate the effects of the intervention on patellofemoral function are lacking. Therefore, a validated and efficient computational model of the patellofemoral joint was developed. The subject specific finite element model was validated against the patellar kinematics recorded during cadaveric patellofemoral laxity experiments of the natural knee. The development involved a sequential process in which the soft-tissue was represented with an increasingly more mechanistic approach with each model iteration. Medial and lateral PF laxity models were developed with the knee at several flexion angles (full extension, ~25 degrees, and ~60 degrees), and a model to simulate passive range of motion was also created. Optimization was conducted to fine-tune a selection of soft-tissue parameters in order to minimize the difference between model-predicted and experimental kinematic results. The average RMS differences for all degrees of freedom and for all flexion angles tested were 2.4 mm and 6.7 degrees with the most simplistic model iteration and 2.5 mm and 5.2 degrees with the most complex model iteration. When the RMS results for medial and lateral PF laxity models are isolated, an improvement is noticed for the iii most complex iteration’s medial laxity results with average RMS differences of 1.6 mm and 4.4 degrees. The validated PF laxity model can be used to assess how changes in knee geometry affect factors such as soft tissue tension and patella tracking.