Domain-informed weight forecasting

Leveraging behavioral and physiological sequences from wearables

Bibliographic Data

ID22074025
AuthorsLuping Cheng (Department of Endocrinology), Lu Wang (0000-0002-9250-3988, Northwestern Polytechnical University), Gaolei Wang (Department of Endocrinology, corresponding author), Bo Lu (0000-0002-3807-7869, Department of Endocrinology, corresponding author), Bei Wu (0000-0002-6891-244X, Northwestern Polytechnical University), Yang Xiao (0000-0001-9237-8913, Department of Endocrinology), Qian Huang (0000-0002-3447-5552, Department of Endocrinology)
Year2026
Volume14
Pages1810381-1810381
Publication date2026-06-02
Open AccessYes
TypeARTICLE
VenueFrontiers in Public Health (JOURNAL)
Journal identifiersISSN: 2296-2565 • E-ISSN: 2296-2565
PublisherFrontiers Media SA (PUBLISHER • CH)
DOI10.3389/fpubh.2026.1810381
PMID42311973
OpenAlexW7163206789
LanguageEN
References cited48

Background: Short-term body-weight forecasting may support personalized weight monitoring, but many existing approaches rely on contemporaneous body weight or body mass index as model inputs, which limits practical use when frequent weigh-ins are unavailable. Methods: We developed a direct multi-step forecasting framework to predict 7-day body-weight trajectories from 14 days of behavioral, physiological, and lifestyle variables. The primary dataset was FitLife360, a synthetic longitudinal dataset on Kaggle, used for model development, benchmarking, and ablation. The LSTM forecaster was evaluated under a participant-level split, with subjects assigned to training, validation, or test sets before window construction. For real-world validation, we tested the framework on PMData from 16 participants over 5 months. Current body weight and BMI were excluded from model inputs and only used as targets. The LSTM was compared with Random Forest, and XGBoost. Results: On the synthetic FitLife360 dataset, the proposed LSTM achieved the best overall performance in the main comparison and showed consistent gains in the feature-ablation analysis. In the supplementary PMData experiment, the same framework remained operational on real-world wearable/lifelogging records, supporting the feasibility of the approach beyond the synthetic development setting. Detailed metrics for the supplementary experiment are reported in the main text. Conclusion: These findings should be interpreted primarily as evidence that domain-informed sequence modeling is feasible for short-horizon body-weight forecasting under both controlled synthetic and supplementary real-world data settings. However, such short-horizon predictions should not be interpreted as direct measures of meaningful adiposity change or short-term cardiometabolic risk, because day-to-day body weight also reflects transient physiological variability. The study therefore provides a methodological foundation for future validation in larger real-world and clinical cohorts

Body mass index · Body weight · Longitudinal data · Synthetic data · Wearable computer · Weight change · Body Composition Measurement Techniques · Mobile Health and mHealth Applications · Non-Invasive Vital Sign Monitoring

Tools

Open DOIOpen Access