Johnathan Bennett; John Shepherd, PhD | University of Hawaii Cancer Center
Competition Sponsor:  National Academy of Medicine
Awardee Year: 2024

Body composition and shape measures, including whole-body and regional fat-free mass, fat mass (e.g., visceral fat), and bone mineral density, are independently associated with health, disease risk, and mortality. However, the development of comprehensive body composition phenotypes and their associations with disease risk remains limited. This project aims to develop risk phenotypes linked to cardiometabolic disease and mortality in large-scale population studies. By leveraging body composition and body shape measures obtained via dual-energy X-ray absorptiometry (DXA), we will identify associations between these parameters and disease risk using advanced machine learning and artificial intelligence models. Our approach combines advanced data science techniques for analyzing DXA scans to derive unique phenotypes with nonlinear statistical modeling to create comprehensive risk prediction models that integrate multiple, synergistic risk factors. This methodology will enable clinical risk modeling at both individual and population levels while uncovering mechanistic insights into the development of specific disease outcomes.