Cluster Analysis of Center-of-Pressure Measures

Abstract

− Postural stability analysis is useful in a variety of applications, such as assessing the risk of falls for older adults or investigating neuromuscular disorders. Variations in the center-of-pressure (COP) are often used to assess postural stability. The COP is a point where the vertical reaction forces of the ground act. It represents the weighted average of all pressures over the body in contact with the ground. Static posturography, which analyses COP variations during a fixed standing posture, is a non-invasive assessment technique, used to quantify postural stability. The objective of this study is to investigate the performance of six conventional COP measures used to characterize postural stability. Six conventional measures are investigated: i) average radial displacement, ii) average velocity, iii) area of the 95% confidence ellipse, iv) standard deviation of COP in the x-direction (medio-lateral), v) standard deviation of COP in the y-direction (antero-posterior), and vi) standard deviation of radial displacement. These COP measures are compared during four different stability conditions: i) feet together, eyes open, ii) feet together, eyes closed, iii) single leg, eyes open, and iv) single leg, eyes closed. Performance is quantified by cluster analysis using the silhouette coefficient, which provides a measure of how well clustered the four stability conditions are for a given stability measure. A good stability measure should have high repeatability for a given stability condition (low intra-cluster distances) and be able to discern between different stability conditions (high inter-cluster distances). Results from eight subjects suggest that out of the six COP measures examined, average velocity is the best measure to assess postural stability. License terms (http://creativecommons.org/licenses/by/2.0). Unrestricted use, distribution, and reproduction in any medium are permitted, provided the original work is properly cited.