Feature selection of stabilometric parameters based on Principal Component Analysis

The present study addresses the open challenge of identifying the features of the Center of Pressure (CoP) trajectory that are most sensitive to postural performance, with the aim of avoiding redundancy and preserving a straightforward interpretation of the results. Postural sway was measured by a force-platform on fifty young healthy subjects. Thirty-seven stabilometric parameters were computed from the one-dimensional and two-dimensional CoP time series. After normalization to relevant biomechanical factors, by means of multiple regression models, a feature selection process was performed based on Principal Component Analysis. Results suggest that CoP two-dimensional time series may be primarily characterized by four parameters, describing: size of the CoP path over the support surface; principal sway direction; shape and bandwidth of the power spectral density plot. CoP one-dimensional time series (antero-posterior, AP, and medio-lateral, ML) may be altogether characterized by six parameters describing: CoP dispersion along AP direction; mean velocity along the ML and AP direction; the contrast between ML and AP regulatory activity; two parameters describing spectral characteristics of CoP along the AP direction. On the basis of the results obtained, some guidelines are suggested on the choice of stabilometric parameters to use, with the view of promoting standardization in quantitative posturography.