Scaling segmental moments of inertia for individual subjects

The purpose of this investigation was to validate methods of scaling human segmental moments of inertia for the transverse principal axis. Firstly, two methods of scaling Chandler et al.'s (Pamphlets DOT HS-801 430 and AMRL TR-74-137, Wright Patterson Air Force Base, OH, 1975) mean subject data to estimate the segmental moments of inertia were used. Chandler et al.'s data were scaled using body mass and segment length (formula 1) or body mass and standing height (formula 2). These data were then compared with a procedure of using the cadaver whose anthropometric measurements most closely match those of the subject. The difference between the criterion data (Chandler's subject data) and scaled values were plotted on scatter diagrams with confidence limits of p less than 0.05 at d.f. = 17. For procedure 1, 43% of the scaled values were plotted within the confidence limits using formula (2) (mass and standing height), compared with 26% for formula (1) (mass and segment length). Formula (1) markedly underestimated the tallest and heaviest subjects. In procedure 2, only 16% and 21% of the scaled values, using formula (1) and (2), respectively, fell within the confidence limits. Results suggested that scaling formulae approximate the moment of inertia of body segments with only limited accuracy. However, if scaling was to be adopted then mean moment of inertia data from an appropriate data set, using the formula that incorporates subject mass and standing height, gave results closest to the criterion value.