A predictive simulation framework for personalised in silico gait retraining in knee osteoarthritis

Area of Science:

• Biomechanics
• Musculoskeletal modeling
• Gait analysis

Background:

• Current gait retraining often uses generic instructions, limiting effectiveness in reducing knee joint loading.
• Generic interventions may not adequately address individual biomechanical needs for optimal knee loading reduction.

Purpose of the Study:

• To develop and demonstrate a proof-of-concept predictive simulation framework for personalized gait retraining.
• The framework explicitly aims to minimize knee joint loading through customized movement patterns.

Main Methods:

• Adapted an open-source framework, enhancing the musculoskeletal model with detailed ligament structures and contact points.
• Updated the objective function to minimize medial/lateral knee joint loading and track patient-specific kinematics.
• Utilized exemplar simulations to test the framework's capabilities.

Main Results:

• The framework successfully tracked lower limb kinematics in the sagittal plane.
• Demonstrated reduction in estimated knee joint contact loading via inverse simulations.
• Showed sensitivity to tibiofemoral alignment and generated novel movement patterns.

Conclusions:

• The developed simulation framework shows promise for personalized gait retraining by reducing knee joint loading.
• In silico results suggest the framework can generate effective movement patterns.
• Further research is needed to assess implementation feasibility and longitudinal effectiveness.