发表机构
Peking University; VITA-Med Group（北京大学; VITA-Med 集团）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出终身数字孪生范式，以事件驱动方式统一终身健康数据，结合智能体框架整合多模态证据，在25个疾病终点上显著提升疾病轨迹预测与风险排序性能。
AI 中文摘要
人类健康是一个连续的、动态的轨迹，由生命历程中生物过程、临床事件、行为和环境暴露的累积相互作用所塑造。统一终身健康信息的全部广度，包括纵向记录、遗传变异、分子谱和环境历史，对于全人建模至关重要，且仍是一个重大挑战。我们提出了终身数字孪生（LifeLong Digital Twin），一种统一的、事件驱动的建模范式，将生活事件（Life Events）组织为每日健康状态（Health States），并将其累积为终身健康上下文（Lifelong Health Context）。配套的智能体框架（Agent Harness）整合了超越语言模型文本上下文的多模态证据。我们在25个疾病终点上评估了四种语言模型，涉及三个任务：疾病轨迹预测、疾病风险排序和多时间跨度疾病预测。该方法相较于参考条件取得了显著提升：在轨迹预测中，疾病识别的模型平均F1提高了22.0%，五年疾病结局的F1提高了18.3%；甲状腺疾病的F1达到0.669。该框架为全人数字孪生和个性化终身疾病预防研究奠定了基础。
英文摘要
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.
Comments15 pages, 5 figures, 2 tables. Technical report
