A regressed phase analysis for coupled joint systems.

Area of Science:

• Biomechanics
• Gait Analysis
• Neurology

Background:

• Relative phase analysis is common for assessing lower limb joint coupling during gait.
• Existing methods have limitations, including reference frame dependency and discontinuities.
• These issues can lead to errors in phase feature extraction and analysis.

Purpose of the Study:

• To address the shortcomings of relative phase analysis in assessing joint coupling.
• To propose and validate an alternative, more robust phase analysis method.
• To improve the accuracy and reliability of gait analysis, particularly for individuals with cerebral palsy.

Main Methods:

• Recorded goniometric data from hip and knee joints of individuals with spastic diplegic cerebral palsy and a healthy control during level walking.
• Applied relative phase analysis and identified its limitations, including reference frame disparity and discontinuities.
• Developed and applied a novel regressed phase analysis comparing joint velocities directly in the velocity-velocity phase plane.

Main Results:

• Relative phase analysis showed significant disparities (16.1±14% gait cycle, 50.6±77% scale) depending on the reference frame (knee-hip vs. hip-knee).
• Discontinuities were observed in relative phase portraits, potentially causing errors in phase feature extraction.
• The proposed regressed phase analysis demonstrated continuity and commutativity between reference frames.

Conclusions:

• The limitations of relative phase analysis stem from its reliance on velocity-position phase plane transformations and the arc-tangent operator.
• Regressed phase analysis offers key advantages, including continuity, commutativity, and generalizability for multi-joint systems.
• This new method provides a more reliable approach to quantifying joint coupling during gait, beneficial for clinical applications in conditions like cerebral palsy.