Longitudinal prediction of BMI using explainable AI: integrating polygenic scores, maternal, early-life and familial factors
The interpretable KAN model revealed the dynamic interplay of childhood BMI z-score and PGS emerging as key drivers of BMI trajectories across life stages. The finding underscores the potential of BMI at critical time in early childhood as a biomarker for obesity risk. Our interpretable model offers actionable insights for targeted obesity prevention strategies.
Authors
- K. Vinsen
- Phillip E. Melton (ORCID: https://orcid.org/0000-0003-4026-2964)
- Rae‐Chi Huang (ORCID: https://orcid.org/0000-0002-8464-6639)
- Trevor A. Mori
- Lawrence J. Beilin
- Fuling Chen
Institutions
- Edith Cowan University (AU)
- University of Tasmania (AU)
- The University of Western Australia (AU)
- International Centre for Radio Astronomy Research (AU)
Publication Details
- Journal
- International Journal of Obesity
- Published
- 2026-03-16
- DOI
- https://doi.org/10.1038/s41366-026-02050-1
- Primary Topic
- Body Composition Measurement Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- University of Notre Dame
- Australian Government
- Raine Medical Research Foundation
- Women and Infants Research Foundation
- Curtin University of Technology
- Edith Cowan University
- Murdoch University
- University of Notre Dame Australia
- Government of Western Australia
- National Health and Medical Research Council