Common functional principal components analysis: a new approach to analyzing human movement data.

Area of Science:

• Biomechanics
• Statistics
• Human Movement Analysis

Background:

• Traditional methods for comparing angle-time series data in human movement studies often discard significant information.
• Existing functional principal components analysis (FPCA) approaches have limitations in directly comparing movement patterns across different groups.
• Group-specific FPCA components can vary in order, hindering cross-group comparisons, while combined FPCA may not represent all groups adequately.

Purpose of the Study:

• To develop and implement a novel statistical technique, common functional principal components analysis (CFPCA), for robust comparison of functional data across multiple groups.
• To address the limitations of existing FPCA methods in analyzing human movement data, particularly for identifying subtle group differences.
• To enable direct and sensible comparisons of movement patterns between groups, enhancing the understanding of biomechanical variations.

Main Methods:

• Functional data analysis (FDA) was employed, treating angle-time series data as continuous functions.
• Common functional principal components analysis (CFPCA) was implemented to identify shared sources of variation across groups while allowing for group-specific component ordering.
• The developed CFPCA method was applied to a biomechanical dataset investigating chronic Achilles tendon injury and orthotic interventions.

Main Results:

• CFPCA successfully identified common sources of variation across different groups in the human movement data.
• The method allowed for direct comparison of functional principal components (FPCs) between groups, overcoming ordering issues.
• Analysis of the Achilles tendon injury dataset demonstrated the utility of CFPCA in discerning functional effects of injury and orthoses.

Conclusions:

• Common functional principal components analysis (CFPCA) provides a powerful and statistically sound approach for comparing functional data across groups in human movement studies.
• CFPCA enhances the ability to detect subtle differences in movement patterns, crucial for understanding injury mechanisms and intervention effects.
• This technique offers a significant advancement over traditional and existing FPCA methods for group comparisons in biomechanics and related fields.