LifeLong Digital Twin: A Unified Modeling Paradigm and Agent Harness for Event-Driven Lifelong Health State Trajectories
Abstract
Human health is a continuous, dynamic trajectory shaped by the cumulative interplay of biological processes, clinical events, behaviors and environmental exposures across the life course. Unifying the full breadth of lifelong health information, including longitudinal records, genetic variation, molecular profiles and environmental histories, is essential for whole-person modeling and remains a major challenge. We introduce LifeLong Digital Twin, a unified, event-driven modeling paradigm that organizes Life Events into daily Health States and accumulates them into Lifelong Health Context. An accompanying Agent Harness incorporates multimodal evidence beyond the language model's textual context. We evaluate four language models across 25 disease endpoints on three tasks: Disease Trajectory Forecasting, Disease Risk Ranking and Multi-horizon Disease Prediction. The approach yields marked gains over the reference condition: model-averaged F1 increases by 22.0% for disease identification in trajectory forecasting and 18.3% for five-year disease outcomes; thyroid-disease F1 reaches 0.669. The framework provides a foundation for whole-person digital twins and research on personalized lifelong disease prevention.
Related
- Oligodendrocyte precursor cell death does not contribute to lifelong population homeostasis
- Lifelong maintenance of locomotion by embryonically active dopamine neurons
- Lifelong Invertebrate Chordate Central Nervous System Regeneration is Driven by a Migrating Pool of Conserved Neural Stem Cells Program
- Hoxa10 governs lifelong skeletal stem cell maintenance and bone repair
- Persistent hippocampal traces of lifelong spatial memory
BibTeX
@misc{jiang2026lifelong,
title = {{LifeLong Digital Twin: A Unified Modeling Paradigm and Agent Harness for Event-Driven Lifelong Health State Trajectories}},
author = {Jin Jiang and Sean Yates and Jasper Chong and Raymond Brooks and Alex Lawson and Yuqin Qin and Liangcai Gao},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.05566},
eprint = {2610.05566},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.05566},
}
RIS
TY - UNPB
TI - LifeLong Digital Twin: A Unified Modeling Paradigm and Agent Harness for Event-Driven Lifelong Health State Trajectories
AU - Jin Jiang
AU - Sean Yates
AU - Jasper Chong
AU - Raymond Brooks
AU - Alex Lawson
AU - Yuqin Qin
AU - Liangcai Gao
PY - 2026
DA - 2026/10/04
PB - arXiv
DO - 10.48550/arxiv.2610.05566
UR - https://arxiv.org/abs/2610.05566
AB - Human health is a continuous, dynamic trajectory shaped by the cumulative interplay of biological processes, clinical events, behaviors and environmental exposures across the life course. Unifying the full breadth of lifelong health information, including longitudinal records, genetic variation, molecular profiles and environmental histories, is essential for whole-person modeling and remains a major challenge. We introduce LifeLong Digital Twin, a unified, event-driven modeling paradigm that organizes Life Events into daily Health States and accumulates them into Lifelong Health Context. An accompanying Agent Harness incorporates multimodal evidence beyond the language model's textual context. We evaluate four language models across 25 disease endpoints on three tasks: Disease Trajectory Forecasting, Disease Risk Ranking and Multi-horizon Disease Prediction. The approach yields marked gains over the reference condition: model-averaged F1 increases by 22.0% for disease identification in trajectory forecasting and 18.3% for five-year disease outcomes; thyroid-disease F1 reaches 0.669. The framework provides a foundation for whole-person digital twins and research on personalized lifelong disease prevention.
ER -