Search Results

You are looking at 1 - 3 of 3 items for

  • Author: Jason Wicke x
Clear All Modify Search
Restricted access

Jason Wicke and Becky Lopers

An elliptical cylinder model developed by Jensen (1978) has been a widely accepted method for determining human segment inertial parameters. The goal of the present study was to evaluate the accuracy of the segment volume calculation step of this model. Three possible sources of error were examined: between-sex differences in body shape, image ratio, and human inconsistencies in digitizing. Volume estimates for the right lower arm + right hand, right lower leg + right foot, and whole body on 20 young men and women were calculated from digitized images at a ratio of 1:10 and 1:5 of the actual size (measured) and compared to values measured using an underwater displacement technique (criterion). Results showed no differences between the sexes on the accuracy of estimating the three volumes at either image ratio. Combining both sexes, the error in calculating segment volumes with an image-to-actual-size ratio of 1:10 were significantly larger, p < 0.05, than at a ratio of 1:5 for both the lower arm + hand (4.28 ± 2.92% vs. –0.43 ± 2.49%) and the whole body (4.80 ± 2.49% vs. 2.01 ± 2.17%). There was no significant change in mean for the lower leg + foot when the image was increased from 1:10 to 1:5 (–0.12 ± 3.92% vs. –0.81 ± 3.01%, respectively). Although not statistically significant, p > 0.05, a greater magnification seemed to have also reduced the influence of human inconsistencies, which was found to be a primary source of error. When the image-to-actual-size ratio is high (i.e., 1:5) and precaution is taken during digitization, the elliptical cylinder model provides accurate estimates of segment volumes of the whole body and extremities.

Restricted access

Jason Wicke and Genevieve A. Dumas

The geometric method combines a volume and a density function to estimate body segment parameters and has the best opportunity for developing the most accurate models. In the trunk, there are many different tissues that greatly differ in density (e.g., bone versus lung). Thus, the density function for the trunk must be particularly sensitive to capture this diversity, such that accurate inertial estimates are possible. Three different models were used to test this hypothesis by estimating trunk inertial parameters of 25 female and 24 male college-aged participants. The outcome of this study indicates that the inertial estimates for the upper and lower trunk are most sensitive to the volume function and not very sensitive to the density function. Although it appears that the uniform density function has a greater influence on inertial estimates in the lower trunk region than in the upper trunk region, this is likely due to the (overestimated) density value used. When geometric models are used to estimate body segment parameters, care must be taken in choosing a model that can accurately estimate segment volumes. Researchers wanting to develop accurate geometric models should focus on the volume function, especially in unique populations (e.g., pregnant or obese individuals).

Restricted access

Jason Wicke and Genevieve A. Dumas

Body segment inertial parameters are required as input parameters when the kinetics of human motion is to be analyzed. However, owing to interindividual differences in body composition, noninvasive inertial estimates are problematic. Dual-energy x-ray absorptiometry (DXA) is a relatively new imaging approach that can provide cost- and time-effective means for estimating these parameters with minimal exposure to radiation. With the introduction of a new generation of DXA machines, utilizing a fan-beam configuration, this study examined their accuracy as well as a new interpolative data-reduction process for estimating inertial parameters. Specifically, the inertial estimates of two objects (an ultra-high molecular density plastic rod and an animal specimen) and 50 participants were obtained. Results showed that the fan-beam DXA, along with the new interpolative data-reduction process, provided highly accurate estimates (0.10–0.39%). A greater variance was observed in the center of mass location and moment of inertia estimates, likely as a result of the course end-point location (1.31 cm). However, using a midpoint interpolation of the end-point locations, errors in the estimates were greatly reduced for the center of mass location (0.64–0.92%) and moments of inertia (–0.23 to –0.48%).