Oral Presentation ANZBMS-MEPSA-ANZORS 2022

Development of predictive statistical shape models for paediatric lower limb bones (#58)

Beichen Shi 1 2 , Martina Barzan 1 2 , Azadeh Nasseri 1 2 , Christopher Carty 1 3 , David Lloyd 1 2 4 , Giorgio Davico 5 6 , Jayishni Maharaj 1 2 , Laura Diamond 1 2 , David Saxby 1 2
  1. Griffith Centre of Biomedical and Rehabilitation Engineering, Menzies Health Institute Queensland, Gold Coast, QLD, Australia
  2. School of Health Sciences and Social Work, Griffith University, Gold Coast, QLD, Australia
  3. Department of Orthopaedic Surgery, Children’s Health Queensland Hospital and Health Service, Brisbane, QLD, Australia
  4. Queensland and Advanced Design and Prototyping Technologies Institute, Griffith University, Gold Coast, QLD, Australia
  5. Department of Industrial Engineering, Alma Mater Studiorum , University of Bologna, Bologna, Italy
  6. Medical Technology Lab, IRCCS Istituto Ortopedico Rizzoli, Bologna, Italy

Background and Objective: Accurate representation of bone shape is important for subject-specific musculoskeletal models as it may influence modelling of joint kinematics, kinetics, and muscle dynamics. Statistical shape modelling is a method to estimate bone shape from minimal information, such as sparse anatomical landmarks, and to avoid the time and cost associated with reconstructing bone shapes from comprehensive medical imaging. Statistical shape models (SSM) of lower limb bones have been developed and validated for adult populations but have limited applicability to paediatric populations. This study aimed to develop SSM for paediatric lower limb bones and evaluate their reconstruction accuracy using sparse anatomical landmarks.

Methods: The SSM for femur, pelvis, tibia, fibula, patella, haunch (i.e., combined femur and pelvis), and shank (i.e., combined tibia and fibula) were generated from manual segmentation of comprehensive magnetic resonance images of 29 typically developing children (15 females; 13±3.5 years) to describe the shape variance of the cohort. We implemented a leave-one-out cross-validation method wherein SSM were used to reconstruct novel bones (i.e., those not included in SSM generation) using sparse-input (i.e., anatomical landmarks), and then compared these reconstructions against bones segmented from magnetic resonance imaging. Reconstruction performance was evaluated using root mean squared errors (RMSE, mm), Jaccard index (0-1), Dice similarity coefficient (DSC) (0-1), and Hausdorff distance (mm). All results reported in this abstract are mean±standard deviation.

Results: Femurs, pelves, tibias, fibulas, and patellae reconstructed via SSM using sparse-input had RMSE ranging from 1.26±0.39 mm (patella) to 3.41±0.95 mm (pelvis), Jaccard indices ranging from 0.61±0.09 (pelvis) to 0.83±0.03 (tibia), DSC ranging from 0.75±0.08 (pelvis) to 0.91±0.02 (tibia), and Hausdorff distances ranging from 3.05±0.90 mm (patella) to 12.41±3.18 mm (pelvis).

Conclusions: The SSM of paediatric lower limb bones showed reconstruction accuracy consistent with previously developed SSM and outperformed adult-based SSM when used to reconstruct paediatric bones.