Related Experiment Video
Updated: Mar 6, 2026
Segmentation and Linear Measurement for Body Composition Analysis using Slice-O-Matic and Horos
Published on: March 21, 2021
Accuracy and Reliability of Assessing Lateral Compartmental Leg Composition Using Dual-Energy X-ray Absorptiometry
Christiana J Raymond1, Tyler A Bosch, Foster K Bush
11Laboratory of Integrative Human Physiology, School of Kinesiology, University of Minnesota, Minneapolis, MN; 2College of Education and Human Development, University of Minnesota, Minneapolis, MN; 3Department of Medicine, University of Minnesota Medical School, University of Minnesota, Minneapolis, MN; and 4Department of Pediatrics, University of Minnesota Medical School, Minneapolis, MN.
Purpose:
This study aimed to investigate the accuracy and reliability of a novel dual-energy x-ray absorptiometry (DXA) scanning method in the frontal plane for total, fat, and lean mass quantification of the anterior and posterior upper leg compartments.
Methods:
Twenty-one (11 females; X¯age = 20.3 ± 1.3 yr) college athletes were assessed for total and regional body composition using DXA. The segmentation of anterior/posterior thigh compartments was measured with participants lying on their right and left sides and the scanned leg elevated with two foam pads. Custom regions of interest (ROIs) were created manually with enCore software for each scan using bony landmarks to quantify lean, fat, and total masses. Paired t-tests assessed this novel positioning method's accuracy against standard positioning. Intraclass correlation coefficients and coefficients of variation examined inter- and intrarater reliability for lateral scan measures of total, fat, and lean masses from manually created ROIs.
Results:
All mean ± SD differences between frontal and lateral DXA scans of right (R) and left (L) leg total mass (R: 8.42 ± 195.57 g; L: 19.47 ± 131.80 g), fat mass (R: 61.26 ± 215.66 g; L: -5.89 ± 239.97 g), and lean mass (R: -103.00 ± 302.54 g; L: -27.58 ± 288.14 g) were nonsignificant (P value range = 0.15-0.91). Intraclass correlation coefficients were high for all composition measures between and within raters, ranging from 0.983 to 0.999 and from 0.954 to 0.999, respectively, with low variation across measures (all coefficients of variation ≤ 5%).
Conclusion:
The results of this study suggest that DXA measures using lateral subject positioning and custom ROIs to assess upper leg total, fat, and lean masses are accurate and reliable compared with total-body frontal subject positioning. Future studies are needed to determine the clinical usefulness of lateral view measures regarding prevention or rehabilitation of sports- or age-related injury.
