Determination of gait patterns in children with spastic diplegic cerebral palsy using principal components

Area of Science:

• Biomedical Engineering
• Clinical Biomechanics
• Pediatric Neurology

Background:

• Cerebral palsy (CP) significantly impacts motor function, particularly gait.
• Spastic diplegia is the most common subtype of CP, characterized by lower limb spasticity.
• Objective gait analysis is crucial for accurate diagnosis and effective treatment planning.

Purpose of the Study:

• To develop an objective, graphical classification method for spastic diplegic cerebral palsy (CP) gait patterns.
• To utilize principal component analysis (PCA) for dimensionality reduction and identification of dominant gait variability.
• To assess the potential of PCA-based graphical classification in clinical settings.

Main Methods:

• Gait analyses were performed on 20 healthy children and 20 children with spastic diplegic CP.
• Principal Component Analysis (PCA) was applied to 27 parameters (26 kinematic variables and age).
• Fuzzy C-mean cluster analysis was used on the principal components to identify distinct gait patterns.

Main Results:

• PCA identified dominant variability, with the first three components explaining 61% of the total variance.
• Healthy children formed a distinct cluster, differentiating them from the CP group.
• Overlapping clusters within the spastic diplegia group allowed for recognition of specific gait patterns.

Conclusions:

• Principal Component Analysis (PCA) provides a quantitative method for classifying cerebral palsy gait types.
• Graphical classification of CP gait patterns can support clinical evaluations and treatment strategies.
• This objective approach can serve as a validation tool for clinical assessments in pediatric CP.