Multiscale musculoskeletal modelling, data-model fusion and electromyography-informed modelling.

Area of Science:

• Biomechanics
• Computational Biology
• Medical Imaging

Background:

• Current musculoskeletal models lack integration across scales.
• Efficient data storage and retrieval for complex biological systems is challenging.
• Personalized modeling requires rapid generation of individual morphology.

Purpose of the Study:

• To advance state-of-the-art musculoskeletal modeling across spatial and temporal scales.
• To develop efficient ontologies and tools for storing and utilizing musculoskeletal models.
• To propose a framework for a clinically relevant virtual physiological human.

Main Methods:

• Population-based modeling for rapid individual morphology generation.
• Multiscale methods for continuum muscle and bone models.
• Mechanostatistical methods (continuum and particle-based) to bridge scales.
• Electromyography-assisted modeling for muscle force prediction.

Main Results:

• Demonstrated population-based modeling for efficient morphology generation from limited data.
• Presented multiscale and mechanostatistical methods for integrated bone and muscle modeling.
• Highlighted the role of muscle in bone remodeling and advanced force prediction techniques.

Conclusions:

• Integrating bone and muscle mechanics is crucial for a clinically relevant virtual physiological human.
• Population-trained models facilitate rapid generation and capture principal modes of muscle and morphology.
• Open-source repositories are essential for community use, personalization, and contribution to musculoskeletal models.