Characterizing and modeling the joint-level variability in human walking.

Area of Science:

• Biomechanics
• Human Movement Analysis
• Gait Dynamics

Background:

• Human gait is typically viewed as periodic, but significant variability exists.
• This variability, distinct from the periodic signal, is crucial for understanding fall risk.
• Previous methods for analyzing gait variability, like step-to-step metrics, do not capture joint-level temporal dynamics.

Purpose of the Study:

• To develop a novel method for accurately capturing joint-level human gait variability over time.
• To model the temporal variations in joint angles during overground walking.
• To establish a framework for analyzing gait variability beyond simple periodic measures.

Main Methods:

• Applied a second-order Fourier series for stance joints and a first-order Fourier series for swing joints to model joint angle variability per step.
• Utilized seven normal distributions, four linear relationships, and twelve continuity constraints to describe inter-step variations of Fourier series parameters.
• Evaluated the method's accuracy in replicating human joint-level variability using qualitative and quantitative analyses with generated curves.

Main Results:

• The proposed Fourier series model accurately captures time-dependent joint-level variability in overground walking at self-selected speeds.
• The inter-step variability of gait can be effectively described using a combination of normal distributions, linear relationships, and continuity constraints.
• Generated curves demonstrated a strong match to human joint-level gait variability patterns.

Conclusions:

• Fourier series provide an effective method for characterizing complex, time-varying joint-level gait variability.
• The developed model offers a more comprehensive approach to gait analysis, potentially improving fall risk assessment.
• This approach advances the understanding of gait dynamics by quantifying variability beyond traditional metrics.