A method for measuring human body composition using digital images

19Citations
Citations of this article
70Readers
Mendeley users who have this article in their library.

Abstract

Background/Objectives Body mass index (BMI) is a proxy for obesity that is commonly used in spite of its limitation in estimating body fatness. Trained observers with repeated exposure to different body types can estimate body fat (BF) of individuals compared to criterion methods with reasonable accuracy. The purpose of this study was to develop and validate a computer algorithm to provide a valid estimate %BF using digital photographs. Subjects/Methods Our sample included 97 children and 226 adults (age in years: 11.3±3.3; 38.1±11.6, respectively). Measured height and weight were used (BMI in kg/m 2 : 20.4±4.4; 28.7±6.6 for children and adults, respectively). Dual x-ray absorptiometry (DXA) was the criterion method. Body volume (BV PHOTO ) and body shape (BS PHOTO ) were derived from two digital images. Final support vector regression (SVR) models were trained using age, sex, race, BMI for % BF NOPHOTO , plus BV PHOTO and BS PHOTO for %BF PHOTO. Separate validation models were used to evaluate the learning algorithm in children and adults. The differences in correlations between %BF DXA , %BF NOPHOTO and %BF PHOTO were tested using the Fisher’s Z-score transformation. Results Mean BF DXA and BF PHOTO were 27.0%±9.2 vs. 26.7%± 7.4 in children and 32.9± 10.4% vs. 32.8%±9.3 in adults. SVR models produced %BF PHOTO values strongly correlated with % BF DXA . Our final model produced correlations of r DP = 0.80 and r DP = 0.87 in children and adults, respectively for %BF PHOTO vs. %BF DXA . The correlation between %BF NOPHOTO and %BF DXA was moderate, yet statistically significant in both children r DB = 0.70; p <0.0001 and adults r DB = 0.86; p<0.0001. However, the correlations for r DP were statistically higher than r DB (%BF DXA vs. %BF NOPHOTO ) in both children and adults (children: Z = 5.95, p<0.001; adults: Z = 3.27, p<0.0001). Conclusions Our photographic method produced valid estimates of BF in both children and adults. Further research is needed to create norms for subgroups by sex, race/ethnicity, and mobility status.

References Powered by Scopus

LIBSVM: A Library for support vector machines

27822Citations
N/AReaders
Get full text

Sarcopenia: An Undiagnosed Condition in Older Adults. Current Consensus Definition: Prevalence, Etiology, and Consequences. International Working Group on Sarcopenia

2423Citations
N/AReaders
Get full text

Comparing correlated correlation coefficients

2128Citations
N/AReaders
Get full text

Cited by Powered by Scopus

Predicting 3D body shape and body composition from conventional 2D photography

24Citations
N/AReaders
Get full text

Agreement of anthropometric and body composition measures predicted from 2D smartphone images and body impedance scales with criterion methods

14Citations
N/AReaders
Get full text

Predicting body measures from 2D images using Convolutional Neural Networks

12Citations
N/AReaders
Get full text

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Cite

CITATION STYLE

APA

Affuso, O., Pradhan, L., Zhang, C., Gao, S., Wiener, H. W., Gower, B., … Allison, D. B. (2018). A method for measuring human body composition using digital images. PLoS ONE, 13(11). https://doi.org/10.1371/journal.pone.0206430

Readers over time

‘18‘19‘20‘21‘22‘23‘2405101520

Readers' Seniority

Tooltip

PhD / Post grad / Masters / Doc 17

57%

Professor / Associate Prof. 6

20%

Researcher 5

17%

Lecturer / Post doc 2

7%

Readers' Discipline

Tooltip

Medicine and Dentistry 8

35%

Nursing and Health Professions 5

22%

Social Sciences 5

22%

Sports and Recreations 5

22%

Save time finding and organizing research with Mendeley

Sign up for free
0