用生理约束神经网络构建糖尿病数字孪生，精准模拟个体血糖变化。
A Physiologically-Constrained Neural Network Digital Twin Framework for Replicating Glucose Dynamics in Type 1 Diabetes
- 基于生理方程约束的神经网络模型，确保可解释性与医学合理性。
- 394个个体数字孪生中，模拟与真实血糖数据在目标范围、低/高血糖时间上等效。
- 适合临床研究者用于个性化治疗仿真与胰岛素优化测试。
模拟1型糖尿病（T1D）患者的葡萄糖动态对开发个性化治疗和支撑数据驱动的临床决策至关重要。现有模型常忽略关键生理机制且难以个性化。本文提出一种生理约束神经网络数字孪生框架，首先构建与葡萄糖调控常微分方程一致的群体级神经网络状态空间模型，并通过形式化验证确保其符合已知的T1D动力学。随后，通过融合个体数据（如血糖管理与情境信息）生成数字孪生，捕捉个体间及个体内的变异性。使用来自T1D运动倡议研究的真实世界数据进行验证：每位参与者两周数据被分割为5小时序列，模拟与观测血糖曲线对比。采用临床等效性检验评估相似性，设定预定义等效边界。在394个数字孪生中，模拟与真实数据在临床指标上等效：血糖目标范围（70–180 mg/dL）占比分别为75.1±21.2%（模拟）与74.4±15.4%（真实；P<0.001）；低于范围（<70 mg/dL）为2.5±5.2% vs. 3.0±3.3%（P=0.022）；高于范围（>180 mg/dL）为22.4±22.0% vs. 22.6±15.9%（P<0.001）。该框架可整合睡眠、活动等未建模因素，同时保持核心动态。该方法支持个性化虚拟治疗测试、胰岛素优化，并融合物理机制与数据驱动建模。代码见：https://github.com/mosqueralopez/T1DSim_AI
原文摘要 · Abstract (English)
Simulating glucose dynamics in individuals with type 1 diabetes (T1D) is critical for developing personalized treatments and supporting data-driven clinical decisions. Existing models often miss key physiological aspects and are difficult to individualize. Here, we introduce physiologically-constrained neural network (NN) digital twins to simulate glucose dynamics in T1D. To ensure interpretability and physiological consistency, we first build a population-level NN state-space model aligned with a set of ordinary differential equations (ODEs) describing glucose regulation. This model is formally verified to conform to known T1D dynamics. Digital twins are then created by augmenting the population model with individual-specific models, which include personal data, such as glucose management and contextual information, capturing both inter- and intra-individual variability. We validate our approach using real-world data from the T1D Exercise Initiative study. Two weeks of data per participant were split into 5-hour sequences and simulated glucose profiles were compared to observed ones. Clinically relevant outcomes were used to assess similarity via paired equivalence t-tests with predefined clinical equivalence margins. Across 394 digital twins, glucose outcomes were equivalent between simulated and observed data: time in range (70-180 mg/dL) was 75.1$\pm$21.2% (simulated) vs. 74.4$\pm$15.4% (real; P<0.001); time below range (<70 mg/dL) 2.5$\pm$5.2% vs. 3.0$\pm$3.3% (P=0.022); and time above range (>180 mg/dL) 22.4$\pm$22.0% vs. 22.6$\pm$15.9% (P<0.001). Our framework can incorporate unmodeled factors like sleep and activity while preserving key dynamics. This approach enables personalized in silico testing of treatments, supports insulin optimization, and integrates physics-based and data-driven modeling. Code: https://github.com/mosqueralopez/T1DSim_AI
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。