Related Experiment Video
Updated: Mar 13, 2026
Author Spotlight: Advancements in 3D Optical Imaging for Comprehensive Body Composition Assessment in Modern Research
Published on: June 7, 2024
Using mobile applications for body composition analysis: A technical review of an artificial intelligence-based tool:
Taiara S Poltronieri1, Bruna R da Silva2, Jonathan Bennett3
1Department of Agricultural, Food & Nutritional Science, University of Alberta, Edmonton, Alberta, Canada; Postgraduate Program in Medical Sciences, Endocrinology, Faculty of Medicine, Federal University of Rio Grande do Sul, Porto Alegre, Rio Grande do Sul, Brazil.
Background:
Advances in health technology have enabled body composition assessments using smartphone photos, offering an accessible, cost-efficient, and portable alternative that can also be used by non-experts. However, it is essential to provide clarity on their technical development and estimation process for clinicians, researchers, and users.
Aim:
Here, we aimed to provide a technical description and guidance on the use and interpretation of a selected artificial intelligence (AI)-based app for body composition estimation.
Methods:
We selected one app as a representative for in-depth technical analysis, based on a non-systematic review of scientific databases, developer websites, search engines, and digital marketplaces, to generate insights relevant to similar tools.
Results:
MeThreeSixty® app was selected due to its availability and validation for several body composition measures (body fat, fat mass, fat-free mass, and appendicular lean mass). The app integrates advanced technologies, such as three-dimensional (3D) imaging and AI, which improves its accuracy with potential for refinement. It also features a self-assessment function to enhance user accessibility. Early findings indicate the app provides reliable group-level results for body circumference and composition estimations, with refinements needed for individual assessments.
Conclusion:
MeThreeSixty app used 3D imaging and AI with acceptable group-level accuracy for estimating body circumference and composition, but limited precision at the individual level requires cautious interpretation. Further prospective validation and model refinement are needed, especially in diverse populations, and using longitudinal datasets before supporting personalized nutrition and broader health platform integration.
