Estimating Body Segment Properties for Adults AcrossDiverse Body Morphologies: A Data-Driven GeometricFramework

Abstract

Accurate estimation of body segment inertial properties is essential for biomechanical analyses, yet commonly used scaling methods rely on limited datasets and do not generalize well across diverse adult body morphologies. We developed a data-driven framework that estimates segment lengths, masses, centers of mass, and moments of inertia using regression models trained on large anthropometric datasets (ANSUR II and NHANES) combined with a geometric representation of 16 body segments. The framework uses height, weight, and sex as primary inputs and incorporates waist and hip circumferences or other length and cross-sectional measurements when available to refine body-shape predictions. For individuals with obesity, additional geometric rules redistribute excess mass based on segment-specific volume changes. The resulting models reproduced segment lengths, cross-sectional dimensions, and lumped segment masses within the ranges observed in the training datasets and outperformed published regression equations, particularly at higher body mass index (BMI) values. To promote broad adoption, we provide an open-source API in Python that performs the full parameter estimation using the trained models. This framework offers an accurate and accessible method for estimating adult body segment properties across a wide range of body sizes and shapes, supporting improved motion analysis, musculoskeletal simulation, and clinical biomechanics.