New method revolutionizes body composition analysis using 3D imaging and deep learning

A recent study introduces an innovative method for analyzing body composition with the help of advanced 3D image formation and deep learning techniques. This approach is intended to offer more accurate reviews of body fat and muscle distribution that are crucial for understanding health risks related to various disorders.

The study, “3D convolutional deep learning for non-linear estimate of the body composition of the entire body morphology,” written by researchers from Pennington Biomedical Research Center, University of Washington, University of Hawaii and University of California-San Francisco was recently published in NPJ Digital MedicineA Journal of the Nature Portfolio.

This study introduces an innovative approach with the help of deep, non -linear methods to improve the estimate of the body composition parameters, thereby surpassing the accuracy of earlier linear models. This progress offers potential for improving assessments in clinical environments and research applications.

Authors of “3D convolutional deep learning for non -linear estimate of body composition from the entire body morphology” include Pennington Biomedical Research Center’s Dr. ir. Steven Heymsfield, Dr. Isaac Tian from the University of Washington, Dr. Jason Liu and Dr. Brian Curless; Dr. Michael Wong from the University of Hawaii, Nisa Kelly, Yong Liu and Dr. John Shepherd; and Dr. Andrea K. Garber from the University of California-San Francisco.

Dr. Steven Heymsfield has extensive experience in human obesity, energy balance regulation and the development of methods for evaluating body composition. His contributions to this area have been crucial in promoting the understanding of human metabolism and the application of new technologies such as 3D optical imaging in medical research. “

Dr. John Kirwan, Pennington Biomedical Executive Director

This development is a step forward in medical imaging and health assessment and offers a more reliable tool for clinics and researchers to evaluate body composition and associated health risks.

“To make a detailed digital map of the body shape of a person easily and quickly and then use that information to not only generate accurate estimates of their body composition and health risks, but also corresponding 3D images, was almost unimaginable just a few years ago , “Said Dr. Heymsfield, professor of metabolism and body composition at Pennington Biomedical. “This kind of technological developments require skills of a wide range of scientists and I am pleased to get the chance here at Pennington Biomedical to collaborate with colleagues from all over the country and the world.”

The most important highlights of the study include:

• Advanced imaging: The researchers used 3D image technology to record detailed representations of the form of the body.
• Deep -Learning Application: By applying advanced deepen algorithms, the study achieved more precise estimates of the body composition compared to traditional methods.
• Health implications: Accurate analysis of body composition is essential for assessing health risks with regard to obesity, cardiovascular disease and other metabolic disorders.

These findings are results of the form up! Studies, financed via AKG: Nichd #R01HD082166, National Institutes of Health Norc Center Grants (P30DK072476, Pennington/Louisiana and P30DK040561, Harvard); Jacket: National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (R01DK109008 and R01DK111698). Visit to read the full article https://www.nature.com/articles/s41746-025-01469-6.