New Advances in Body Composition Assessment with ShapedNet: A Single Image Deep Regression Approach

TL;DR

ShapedNet is a deep learning model that accurately estimates body fat percentage, identifies individuals, and localizes features from a single image, outperforming existing computer vision methods and validated against DXA standards.

Contribution

This paper introduces ShapedNet, a novel multi-task deep neural network for body composition assessment from a single image, advancing the field's accuracy and efficiency.

Findings

ShapedNet achieves a MAPE of 4.91% in body fat estimation.

Outperforms 19.5% of state-of-the-art methods.

Validated on 1273 diverse adults against DXA.

Abstract

We introduce a novel technique called ShapedNet to enhance body composition assessment. This method employs a deep neural network capable of estimating Body Fat Percentage (BFP), performing individual identification, and enabling localization using a single photograph. The accuracy of ShapedNet is validated through comprehensive comparisons against the gold standard method, Dual-Energy X-ray Absorptiometry (DXA), utilizing 1273 healthy adults spanning various ages, sexes, and BFP levels. The results demonstrate that ShapedNet outperforms in 19.5% state of the art computer vision-based approaches for body fat estimation, achieving a Mean Absolute Percentage Error (MAPE) of 4.91% and Mean Absolute Error (MAE) of 1.42. The study evaluates both gender-based and Gender-neutral approaches, with the latter showcasing superior performance. The method estimates BFP with 95% confidence within an error margin of 4.01% to 5.81%. This research advances multi-task learning and body composition assessment theory through ShapedNet.