Personalization of Closed-Chain Shoulder Models Yields High Kinematic Accuracy for Multiple Motions

Area of Science:

• Biomechanics
• Musculoskeletal modeling
• Orthopedics

Background:

• Shoulder joint complex issues, like rotator cuff pain, affect many adults with poor treatment outcomes.
• Current musculoskeletal models lack personalization and accuracy for shoulder biomechanics.
• Limitations hinder precise prediction of joint/muscle loads and simulation of anatomical motions.

Purpose of the Study:

• Develop a novel, personalized modeling framework for the shoulder complex.
• Calibrate subject-specific joint centers and functional axes.
• Enhance accuracy in shoulder biomechanical simulations.

Main Methods:

• Utilized in vivo biplane fluoroscopy data and the Joint Model Personalization Tool.
• Optimized joint parameters and body scale factors for shoulder models with varying degrees of freedom (DOFs).
• Developed and tested open-chain and closed-chain shoulder models (3-5 DOFs).

Main Results:

• Increasing DOFs in open-chain models improved accuracy, with 5-DOF models showing lowest errors (avg. 0.8 mm).
• Closed-chain shoulder models with 5-DOF scapula achieved high accuracy (avg. 0.9 mm) and consistent performance across subjects.
• The framework demonstrated minimized errors in joint kinematics.

Conclusions:

• The developed personalized modeling framework enhances shoulder biomechanical simulation accuracy.
• This approach minimizes joint kinematics errors, providing a foundation for future personalized models.
• Potential clinical utility for improved rehabilitation and surgical planning in shoulder conditions.