Feasibility of a Center of Mass Based Fuzzy-Logic Phase Detection Algorithm for Post-Spinal Cord Injury Gait

Area of Science:

• Biomechanics
• Neurorehabilitation
• Computational Intelligence

Background:

• Current spinal cord injury (SCI) walking restoration methods using neuromuscular stimulation often result in abnormal gait patterns.
• Center of Mass (CoM) trajectories during neurotypical walking offer insights into whole-body movement that could improve gait control.
• Fuzzy Logic Algorithms (FLA) are suitable for processing noisy biological data and mimicking human reasoning.

Purpose of the Study:

• To assess the feasibility of a CoM-based FLA for accurate gait phase detection.
• To determine the potential of this FLA for controlling stimulation to enhance walking mechanics post-SCI.
• To investigate the application of CoM kinematics in gait phase detection for individuals with SCI.

Main Methods:

• Developed a CoM-based FLA using five inputs derived from overground walking data of neurotypical subjects.
• Optimized the FLA using a Genetics Algorithm and performed cross-validation.
• Verified the FLA offline using data from neurotypical individuals and individuals with SCI, calculating goodness indices (G).

Main Results:

• The FLA demonstrated good performance in detecting double support and swing phases (G≤0.52) across all participants.
• The study successfully validated the CoM-based FLA's capability for gait phase detection.
• The findings indicate that CoM kinematics can be effectively utilized as features for gait phase identification.

Conclusions:

• Utilizing Center of Mass (CoM) kinematics is a feasible approach for gait phase detection in individuals with spinal cord injury (SCI).
• CoM-based Fuzzy Logic Algorithms (FLA) show potential for improving the control of neuromuscular stimulation to restore more natural walking patterns post-SCI.
• This research supports the development of advanced neurorehabilitation strategies leveraging biomechanical data for enhanced gait recovery.