(Dr.-Ing. Malte Asseln, Assistant Professor, University of Twente, 08. May, 2025 )
Total knee arthroplasty (TKA) is a successful surgical procedure that provides pain relief and restores patient mobility. Despite encouraging implant survival rates, patient satisfaction is stagnant and up to 30% of patients are dissatisfied due to limited function associated with activities of daily living. Currently, only static biomechanical approaches are considered in the clinical workflow. For example, implant alignment is based on a geometric load axis at full leg extension, and the intraoperative functional testing and ligament balancing are performed by manual manipulation by the surgeon. Up to 90% of the forces in the knee joint are not present in the unloaded intraoperative situation. Personalized musculoskeletal models have the potential to predict the active interactions of the knee joint and provide valuable insights for functional optimization of implant alignment or for personalized implant design. The development of a personalized musculoskeletal model for total knee arthroplasty is a challenging task, especially given the limited data available in routine clinical practice. Typically, anthropometric data and medical imaging data of the knee joint region, hip and ankle are acquired. However, inverse dynamic analysis requires motion data and external loads, which are available in a research environment but not in general practice.
In this webcast, we present a personalized musculoskeletal model suitable for predicting the TKA mechanics based on clinical input data using the AnyBody Modeling System. Rather than personalizing a full body model, we built a lower extremity model from scratch based solely on input data available in routine clinical practice. A squat motion was simulated based on a statistical description, external loads were related to anthropometry, and bony landmarks and soft tissue attachments were predicted based on annotated statistical shape models. The approach was validated using in vivo force measurements. The model showed comparable accuracy to previously published musculoskeletal models in estimating in vivo knee contact forces, but with a minimum of input data and reduced personalization effort.
Presented by: Dr.-Ing. Malte Asseln, Assistant Professor, University of Twente, Department of Biomechanical Engineering, Chair of Medical Device Design & Production, Enschede, The Netherlands (current position) Chair of Medical Engineering at the Helmholtz-Institute for Biomedical Engineering at RWTH Aachen University, Aachen, Germany (PhD institution from where the work for this webcast essentially originates)
Playback (154 MB)YouTube