From body hulls to musculoskeletal models: Personalized inertial parameter estimation

Area of Science:

• Biomechanics
• Human Movement Analysis
• Musculoskeletal Modeling

Background:

• Current musculoskeletal models rely on generic templates scaled by body mass and segment lengths, neglecting individual body shapes.
• This scaling approach can introduce significant errors in estimating joint forces and torques, crucial for accurate musculoskeletal analysis.
• Existing methods lack personalization, limiting the precision of movement analysis in diverse populations.

Purpose of the Study:

• To develop and validate a novel method for creating participant-specific musculoskeletal models using smartphone-acquired body hulls.
• To improve the accuracy of estimating body segment inertial parameters by accounting for individual body shapes.
• To introduce new generic musculoskeletal models for broader applicability and comparison.

Main Methods:

• Estimating body segment inertial parameters from 3D body hulls captured via smartphone images.
• Inferring skeletal shape and pose, then estimating tissue distribution (bone, lean, fat) and density.
• Segmenting the body hull to assign tissue-specific densities for inertial parameter calculation.
• Developing two new generic musculoskeletal models based on average body shapes for comparison.

Main Results:

• Participant-specific models derived from the new method showed higher accuracy compared to scaled generic models when validated against MRI data.
• Lab-based gait analysis demonstrated a reduction in residual forces by up to 14.9% using the personalized models.
• A significant reduction in metabolic cost, up to 12.8%, was observed with the personalized musculoskeletal models.
• The new generic models yielded comparable joint moment results but offered less reduction in residual forces than personalized models.

Conclusions:

• The developed method effectively generates accurate, participant-specific musculoskeletal models from smartphone data.
• Personalized musculoskeletal models significantly improve the precision of movement analysis, reducing errors in force and moment estimations.
• This approach offers a practical and accessible way to enhance the accuracy of biomechanical studies and clinical applications.