Cluster analysis for the extraction of sagittal gait patterns in children with cerebral palsy

Area of Science:

• Biomechanical analysis
• Pediatric orthopedics
• Movement science

Background:

• Standardizing gait disorder classification is crucial for consistent management and interdisciplinary communication.
• Previous classification methods lacked clear methodology and validation.
• Gait pattern variability in children necessitates a robust classification system.

Purpose of the Study:

• To apply hierarchical cluster analysis to sagittal kinematic gait data from children.
• To define and validate distinct clusters of gait patterns in pediatric populations.
• To establish a standardized framework for gait disorder classification.

Main Methods:

• Hierarchical cluster analysis was performed on gait data from 56 children with cerebral palsy and 11 controls.
• A structured rationale was used to determine the optimal number of homogenous gait types.
• Gait data included sagittal kinematic parameters.
• Visual assessment and a structured protocol aided in validating gait groupings.

Main Results:

• Thirteen distinct gait clusters were identified within the dataset.
• These clusters were organized into three main categories: 'crouch gait type', 'equinus gait type', and 'other gait type'.
• The cluster analysis successfully defined valid and distinct gait groupings.

Conclusions:

• Hierarchical cluster analysis provides a valid method for classifying pediatric gait disorders.
• The identified gait clusters offer a standardized approach to understanding and managing gait patterns.
• This classification system has the potential to enhance communication and care for children with gait abnormalities.