Comparison of a mobile application to estimate percentage body fat to other non-laboratory based measurements

Abstract

SummaryStudy aim: The measurement of body composition is important from a population perspective as it is a variable associated with a person's health, and also from a sporting perspective as it can be used to evaluate training. This study aimed to examine the reliability of a mobile application that estimates body composition by digitising a two-dimensional image. Materials and methods: Thirty participants (15 men and 15 women) volunteered to have their percentage body fat (%BF) estimated via three different methods (skinfold measurements, SFM; bio-electrical impedance, BIA; LeanScreen TM mobile application, LSA). Intra-method reproducibility was assessed using intra-class correlation coefficients (ICC), coefficient of variance (CV) and typical error of measurement (TEM). The average measurement for each method were also compared. Results: There were no significant differences between the methods for estimated %BF (p = 0.818) and the reliability of each method as assessed via ICC was good (≥0.974). However the absolute reproducibility, as measured by CV and TEM, was much higher in SFM and BIA (≤1.07 and ≤0.37 respectively) compared with LSA (CV 6.47, TEM 1.6). Conclusion: LSA may offer an alternative to other field-based measures for practitioners, however individual variance should be considered to develop an understanding of minimal worthwhile change, as it may not be suitable for a one-off measurement.

Smartphone and Tablet Software Apps to Collect Data in Sport and Exercise Settings: Cross-sectional International Survey

“…For example, there is some inconsistency in the use of software apps to measure heart rate [8], particularly during exercise of various intensities [9]. Furthermore, a recent review by Peart et al [10] highlighted inconsistencies in validity and reliability when estimating body fat percentage using a range of commercially available software [11][12][13].…”

Mobile applications for the sport and exercise nutritionist: a narrative review

“…For example, LeanScreen (Postureco, Trinity, FL, USA) is a smartphone app that estimates percentage body fat by digitizing a series of girths within 2-dimensional (2D) photographs. This reliability of the app has been supported, with inter- and intrareliability coefficients of ≥ 0.99 in a study from MacDonald et al [ 54 ], and an intratester coefficient of 0.974 in another study [ 55 ]. Of note in the latter study though is that typical error of measurement (TEM), a measure of within-subject variation calculated as the standard deviation of repeated measurements, was higher in the app (TEM = 1.6%BF) when compared to skinfold calipers (TEM = 0.37%BF) and bioelectrical impedance analysis (TEM = 0.23%BF).…”

Comparison of a 2D iPad application and 3D body scanner to air displacement plethysmography for measurement of body fat percentage

“…Shaw et al . reported no significant differences when comparing estimates of %BF obtained from the 2D APP to estimates obtained from other field methods. The app‐estimated mean (SD) of 21.9 (6.7)%BF was within ±1%BF of the skinfold [22.9 (4.8)%BF] and bioimpedance [22.3 (7.6)%BF] measures in a sample of 15 males and 15 females.…”

“…Similarly, Shaw et al . reported high relative test–retest reliability (ICC = 0.974) for the LeanScreen app, although they cautioned that the absolute reliability (coefficient of variation = 6.5%) was not as good as the skinfold and bioimpedance methods. Comparatively, the coefficient of variation for %BF estimations from the Bod Pod is 3.1% .…”