Predicting Anthropometric Body Composition Variables Using 3D Optical Imaging and Machine Learning

Abstract

Accurate prediction of anthropometric body composition variables, such as Appendicular Lean Mass (ALM), Body Fat Percentage (BFP), and Bone Mineral Density (BMD), is essential for early diagnosis of several chronic diseases. Currently, researchers rely on Dual-Energy X-ray Absorptiometry (DXA) scans to measure these metrics; however, DXA scans are costly and time-consuming. This work proposes an alternative to DXA scans by applying statistical and machine learning models on biomarkers (height, volume, left calf circumference, etc) obtained from 3D optical images. The dataset consists of 847 patients and was sourced from the Penning-ton Biomedical Research Center. Extracting patients' data in healthcare faces many technical challenges and legal restrictions. However, most supervised machine learning algorithms are inherently data-intensive, requiring a large amount of training data. To address this challenge, we compare the standard supervised to a semi-supervised p-Laplacian model, which leverages the limited data by incorporating the unlabeled patient information. To our knowledge, this paper is the first to demonstrate the application of a game-theoretic p-Laplacian model for regression in healthcare. Our p-Laplacian model yielded errors of ∼13% for ALM, ∼10% for BMD, and ∼20% for BFP when the training data accounted for 10 percent of all data. Among the supervised algorithms we implemented, Support Vector Regression (SVR) performed the best for ALM and BMD, yielding errors of ∼8% for both, whereas Least Squares SVR performed the best for BFP with ∼11% error when trained on 80% the data. Our findings position the p-Laplacian model as a promising tool for healthcare applications, particularly in a data-constrained environment with limited labeled data.