Myobolica: A Stochastic Approach to Estimate Physiological Muscle Control Variability

Area of Science:

• Biomechanics
• Musculoskeletal Modeling
• Computational Biology

Background:

• Traditional musculoskeletal models use cost function optimization, which may not accurately represent non-adult or pathological populations.
• Stochastic approaches, like Metabolica, sample multiple solutions but can allow abrupt muscle activation changes.
• Existing methods face limitations in accurately modeling diverse populations and physiological constraints.

Purpose of the Study:

• To evaluate a new stochastic approach, Myobolica, for musculoskeletal modeling.
• To compare Myobolica's muscle control variability (solution band width) against the Metabolica model.
• To assess the physiological relevance of Myobolica by comparing its solutions to experimental data and static optimization.

Main Methods:

• Developed Myobolica, an enhanced stochastic approach incorporating physiological constraints and parameters.
• Compared muscle force and knee joint force solution bands generated by Myobolica and Metabolica.
• Validated solutions against results from static optimization and experimentally measured knee joint forces.

Main Results:

• Myobolica significantly narrowed the solution band compared to Metabolica, indicating reduced muscle control variability.
• Myobolica solutions demonstrated strong correlation with experimental data (R² = 0.92, RMSE = 0.3 BW).
• Myobolica solutions showed a moderate correlation with the static optimization solution (R² = 0.82, RMSE = 0.63 BW).

Conclusions:

• Myobolica provides a more physiologically relevant and constrained set of solutions for musculoskeletal modeling.
• The refined approach offers improved muscle control variability estimation compared to previous stochastic methods.
• Further research is needed to confirm findings and enhance the Myobolica implementation.