Patient-specific musculoskeletal models can provide accurate diagnostic information and have the potential to improve the treatment of musculoskeletal disorders, by helping surgeons formulate optimal pre-operative planning for each patient. In this webcast, we present an award-winning patient-specific musculoskeletal model. It is capable of estimating muscle, ligament, joint contact forces and knee kinematics simultaneously. We show how it is used to investigate component size and alignment problems in total knee arthroplasty. In addition, we demonstrate techniques to accelerate the estimation of the joint contact forces using artificial neural networks, and how to embed an efficient elastic foundation contact model for the analysis of patellofemoral contact pressure in a model of trochlear dysplasia. This research is funded by the ERC under grant agreement no. 323091.