Estimating model parameters of a mathematical model for hand motion during cylindrical grasps

Area of Science:

• Biomechanics
• Human Motion Analysis
• Mathematical Modeling

Background:

• Understanding human finger biomechanics is crucial for prosthetics, rehabilitation, and robotics.
• Accurate musculoskeletal models are needed to simulate and predict finger movements.
• Existing models may not fully capture the complex dynamics of passive moments and muscle scaling.

Purpose of the Study:

• To develop and validate a novel musculoskeletal model for human finger segments.
• To estimate passive moment and musculoskeletal scaling coefficients using a least squares optimization approach.
• To assess the model's ability to accurately represent individual finger kinematics.

Main Methods:

• Utilized a novel musculoskeletal model as the actuator system.
• Employed least squares optimization to estimate model parameters.
• Compared model-predicted angular data with motion capture system data.

Main Results:

• Achieved a high degree of agreement between model predictions and filtered motion capture data (R² = 0.98).
• Parameter estimates were found to be unique to the individual and the specific dataset.
• Demonstrated the model's capability in capturing the dynamics of human finger motion.

Conclusions:

• The developed mathematical model and parameter estimation approach provide accurate insights into human finger biomechanics.
• The model successfully determined unique passive moment and musculoskeletal scaling coefficients.
• This methodology holds potential for application across diverse grasp types and in personalized biomechanical analysis.