Article: Automatic generation of personalised human body 3D models using a parametric model with a statistical model to infer body composition from body mass index Journal: International Journal of Human Factors Modelling and Simulation (IJHFMS) 2026 Vol.8 No.2 pp.145 - 168 Abstract: Personalised 3D models of patients' bodies constitute an effective approach for rapidly and accurately evaluating burned and total skin surfaces in constrained burn care settings. This study aims to develop an automatic method for generating 3D models of human subjects with accurate body shapes from accessible data (gender, age, height, weight). We developed a statistical model to infer three body composition parameters from the body mass index (BMI) to adjust the weight, muscle and micro ratios in MakeHuman, defining the subject's global and local morphologies. The proposed approach improves body surface area estimation, with a mean percentage error of 2.10%, outperforming the results of traditional anthropometric formulas and parametric modelling methods used in clinical settings. Across the population, it provides a continuous representation of human morphology, covering a wide range of body shapes and providing a valuable tool for burn assessment and personalised care applications, particularly for obese individuals. Inderscience Publishers - linking academia, business and industry through research

Title: Automatic generation of personalised human body 3D models using a parametric model with a statistical model to infer body composition from body mass index

Authors: Anaïde Thibault; Thierry Cresson; Jacques A. de Guise; Carlos Vázquez

Addresses: École de Technologie Supérieure (ÉTS), 1100 rue Notre Dame Ouest, Montréal, Québec H3C 1K3, Canada ' École de Technologie Supérieure (ÉTS), 1100 rue Notre Dame Ouest, Montréal, Québec H3C 1K3, Canada ' École de Technologie Supérieure (ÉTS), 1100 rue Notre Dame Ouest, Montréal, Québec H3C 1K3, Canada ' École de Technologie Supérieure (ÉTS), 1100 rue Notre Dame Ouest, Montréal, Québec H3C 1K3, Canada

Abstract: Personalised 3D models of patients' bodies constitute an effective approach for rapidly and accurately evaluating burned and total skin surfaces in constrained burn care settings. This study aims to develop an automatic method for generating 3D models of human subjects with accurate body shapes from accessible data (gender, age, height, weight). We developed a statistical model to infer three body composition parameters from the body mass index (BMI) to adjust the weight, muscle and micro ratios in MakeHuman, defining the subject's global and local morphologies. The proposed approach improves body surface area estimation, with a mean percentage error of 2.10%, outperforming the results of traditional anthropometric formulas and parametric modelling methods used in clinical settings. Across the population, it provides a continuous representation of human morphology, covering a wide range of body shapes and providing a valuable tool for burn assessment and personalised care applications, particularly for obese individuals.

Keywords: human body 3D modelling; body shapes; parametric modelling; MakeHuman; statistical model; automation; body mass index; BMI.

DOI: 10.1504/IJHFMS.2026.153490

International Journal of Human Factors Modelling and Simulation, 2026 Vol.8 No.2, pp.145 - 168

Received: 19 Dec 2024
Accepted: 16 Dec 2025
Published online: 11 May 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Automatic generation of personalised human body 3D models using a parametric model with a statistical model to infer body composition from body mass index

Keep up-to-date