Data-driven spectral analysis for coordinative structures in periodic human locomotion

Area of Science:

• Nonlinear dynamics
• Systems biology
• Biomechanics

Background:

• Living organisms utilize complex, redundant control mechanisms involving numerous components.
• Conventional methods for extracting coordinative structures often fail to capture essential dynamical properties.
• Understanding these structures is crucial for analyzing biological periodic systems.

Purpose of the Study:

• To develop a data-driven method for analyzing coordinative structures in biological periodic systems.
• To extract dynamical properties like frequency and phase from these structures.
• To apply and validate the approach using human locomotion and simulation data.

Main Methods:

• Modeling biological periodic systems as nonlinear limit-cycle oscillations.
• Applying operator-theoretic spectral analysis to extract dynamical properties.
• Utilizing segmental angle series from human walking and simulation data.

Main Results:

• Identified speed-independent coordinative structures related to gait frequency harmonics.
• Discovered speed-dependent, time-evolving phase behaviors using estimated eigenfunctions.
• Validated the approach with double pendulum and walking model simulations.

Conclusions:

• The proposed operator-theoretic spectral analysis effectively captures dynamical properties of biological coordinative structures.
• This method provides a new perspective for analyzing biological periodic phenomena through nonlinear dynamical systems.
• The findings have implications for understanding locomotion and other periodic biological processes.