An entropy-assisted musculoskeletal shoulder model.

Area of Science:

• Biomechanics
• Musculoskeletal modeling
• Human movement analysis

Background:

• Musculoskeletal shoulder models estimate mechanical loading using optimization.
• Traditional models minimize muscle activity but neglect antagonist co-contraction.
• Antagonist co-contraction is crucial for shoulder joint stability and control.

Purpose of the Study:

• To propose a novel objective function incorporating an entropy term to account for muscle co-contractions.
• To develop and validate a musculoskeletal shoulder model using the proposed objective function.
• To investigate the impact of the entropy term's weight on model predictability.

Main Methods:

• Developed a musculoskeletal shoulder model with an entropy-based objective function.
• Conducted experiments with participants generating 3D shoulder moments in various postures.
• Measured surface electromyography (EMG) of 8 shoulder muscles.
• Compared model predictions with experimental EMG data using Bhattacharyya distance and concordance ratio.

Main Results:

• A small weight for the entropy term significantly improved the model's predictability of muscle activities.
• The proposed objective function better captures muscle co-contraction dynamics compared to traditional methods.
• Experimental validation confirmed the enhanced accuracy of the entropy-inclusive model.

Conclusions:

• Incorporating an entropy term into musculoskeletal shoulder models enhances their validity.
• Entropy offers a valuable concept for understanding muscle co-contraction mechanisms.
• This approach advances the development of more accurate shoulder biomechanical models.