Detecting dynamical boundaries from kinematic data in biomechanics.

Area of Science:

• Dynamical systems theory
• Biomechanics
• Nonlinear dynamics

Background:

• Dynamical boundaries are crucial for understanding system stability.
• Current methods require phase space vector fields, limiting applicability.
• Musculoskeletal biomechanics presents complex systems with separatrix features.

Purpose of the Study:

• To develop a method for identifying dynamical boundaries using only reconstructed time series data.
• To apply this method to problems in musculoskeletal biomechanics, specifically postural control and balance.
• To establish novel measures for stability and fall risk.

Main Methods:

• Utilizing ridges in the state space distribution of finite-time Lyapunov exponents.
• Reconstructing trajectories from time series data.
• Analyzing human balance data and models.

Main Results:

• Successfully determined dynamical boundaries in human balance activities.
• Demonstrated the ability to locate boundaries without requiring a full phase space vector field.
• Identified the boundary between recovery and failure in balance tasks.

Conclusions:

• The developed method provides a robust approach to dynamical boundary detection from time series.
• This technique offers new, currently unavailable, measures for stability and fall risk.
• Potential benefits include improved analysis and prevention of low back pain and fall-related injuries.