Abstract Summary
Commonly used musculoskeletal models rely on scaled generic bony geometries, which poorly represent an individual’s skeletal anatomy. This results in erroneous joint centre location, muscle origin and insertion points, and muscle-tendon moment arm estimates leading to inaccurate biomechanical analyses. Personalised models based on motion capture and medical imaging data can be created using open-source statistical shape modelling software, such as the MAP Client. To date, the MAP Client has not been applied to paediatric populations and therefore, this study aimed to determine whether accurate pelvis, femur, and tibiofibular paediatric bones could be accurately reconstructed combining statistical shape modelling and morphing techniques.