Effects of foot orthoses on the musculoskeletal system of recreational runners: a functional data analysis approach
The core objective for this project was to explore the possibility of using Functional Principal Component Analysis (FPCA) to discriminate between four shoe orthoses for different measured variables. Before doing this, a summary of the measured data was considered in chapter 3. This study recognized differences in the mean and standard deviation functions representing all four considered variables under the different conditions. A look at the autocorrelations between the variables at different time points, using contour plots and sometimes perspective plots under every condition of shoe orthosis also indicated significant differences between the various conditions and their effect on the autocorrelations at different points in the stance phase. Notable among them is the difference between the correlation structure of foot eversión under control and medial posting orthosis. Here the foot of subjects in the first half of the stance phase appeared to be in a resting phase. The scatter plots of most bivariate/multivariate principal components showed some clustering. This can be explained by the fact that there were several trials by same subjects over different sessions, so similar P C scores might have been obtained for similar curves and thus leading to the clustering. Bivariate and multivariate FPCA indicated that foot eversión velocity dominates all other variables under almost all conditions with respect to their individual contribution to the variability represented by PCl and PC2. For scaled bivariate FPCA, there was a balance between all variables involved, as to their contribution to the variability represented by PCl and PC2.
Bibliography: p. 137-140
Amuah, J. E. (2004). Effects of foot orthoses on the musculoskeletal system of recreational runners: a functional data analysis approach (Master's thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/15175