Modelling the Form and Function of the Paediatric Musculoskeletal System

Abstract

Musculoskeletal development in a paediatric population is an important but poorly understood domain. Understanding how bone geometry changes with growth, allows development of more accurate models in a paediatric population. Which can be used for applications such as musculoskeletal and finite element modelling, to help understand information about movement including joint contact and muscle forces. Utilising a dataset of 333 CT scans of children aged 4 – 18 years old, this thesis explores changes in musculoskeletal geometry with growth. This thesis evaluates clinical bone measurement changes with growth. Three-dimensional angular and linear measurements from the pelvis, femur, and tibia/fibula were extracted. Notably, femoral and tibial torsion remained consistent between sexes, while sex-related differences in linear measurements emerged in children aged 13 and older. Females consistently displayed smaller linear measurements compared to males of the same age/height, possibly contributing to increased injury risks during adolescence. Accurate estimation of the hip joint centre is vital for assessing hip joint motion and related parameters. In this thesis, three different hip joint centre determination methods were evaluated in a paediatric population—regression equations, linear scaling, and shape model prediction. Our novel shape model prediction exhibited superior accuracy compared to alternative methods. Addressing the challenge of rapidly generating musculoskeletal geometry in a paediatric population, this thesis presents population-based shape models for lower limb bones. These models utilised principal component analysis on segmented pelvis, femur, and tibia/fibula data, to accurately predict bone geometry from demographic and linear bone measurements. Subsequent articulation of the shape model allowed for both shape and pose prediction of lower limb bones from bone surface landmarks. The generated shape models outperformed traditional linear scaling methods, offering a more accurate and accessible solution for paediatric musculoskeletal geometry prediction. Overall, this thesis underscores the importance of understanding changes in musculoskeletal geometry with growth and provides valuable tools and methods which capture these changes. This thesis not only illustrates sex-related skeletal differences but has also pioneered the use of statistical shape models in a paediatric population to predict lower limb musculoskeletal geometry. These findings hold significant implications for clinical and research applications, advancing our understanding of paediatric musculoskeletal geometry.