Understanding the variation of tibia and femur bone mineral density in a paediatric population

Unlike FE models based on adult bone geometry, limited literature is available for paediatric FE modelling. For example, paediatric bone mineral density (BMD) and therefore bone material properties vary with age, the presence of growth plate and cortical thickness. The influence of these parameters have not been studied to date and literature is lacking the required dataset. We have an existing dataset of 333 calibrated CT scans of children aged 4 to 18 years allowing us to extract bone mineral density information for the tibia/fibula, femur and pelvis. The first step will be to create a statistical BMD model of the growing child bone mineral density using machine-learning algorithms. The second step will be to understand the influence of bone mineral density and cortical thickness on the femur and tibia Von Mises stress distribution. The third aim will be to build the first personalised paediatric knee FE model.