A three-dimensional regression model of the shoulder rhythm

Area of Science:

• Biomechanics
• Orthopedics
• Human Movement Science

Background:

• Current 3D shoulder motion analysis is lab-bound, limiting real-world applications in ergonomics and sports.
• Consistent relationships exist between scapular and humeral movements, suggesting potential for predictive modeling.
• A descriptive statistical model of shoulder rhythm is needed for in-situ biomechanical research.

Purpose of the Study:

• To develop a thorax-fixed regression model for the shoulder.
• To statistically predict clavicle and scapula orientation from humerus orientation.
• To enable in-situ biomechanical analysis of the shoulder mechanism.

Main Methods:

• Recorded shoulder bone orientations for 10 subjects across 23 humerus positions.
• Applied a 20 N external force at the elbow during arm abduction/adduction with minimal other forces.
• Used repeated measures multivariate analysis of variance to determine linear regression equations for clavicular and scapular orientations.

Main Results:

• Five linear regression equations accurately predicted clavicle and scapula orientation.
• Key predictors included humerus orientation, external force direction, and initial position.
• Subject morphology and gender did not significantly impact prediction accuracy.

Conclusions:

• A validated statistical model predicts clavicular and scapular orientations based on humerus position, initial posture, and external force.
• The model demonstrates good fit with independent data from a new subject group.
• The model provides a non-invasive method for assessing shoulder kinematics in diverse settings.