文献速递医学人工智能与信息学流行病学与生物统计基础与转化研究

基于纵向电子健康记录和母子AI智能体预测母婴结局

Prediction of maternal and infant outcomes from longitudinal electronic health records with a mother-child AI agent.
PubMed 文献 · Nat Med · 2026-09-04

研究开发了基于大语言模型的母子人工智能智能体MoChiAgent,其核心预测引擎MoChiFormer利用4,401,599次纵向临床就诊数据开发,并在263,452和23,192次就诊的独立母婴队列中外部验证。MoChiFormer可重建缺失实验室值、减少批次效应,准确识别关键妊娠疾病,并揭示跨代风险关联。

要点

  • MoChiFormer对胎盘早剥、胎膜早破和早产的AUC分别为0.89、0.89和0.91。
  • 特定母亲集群所生婴儿的新生儿黄疸风险比为2.81(95%CI 2.60-3.03)。
  • 特定母亲集群所生婴儿的血液疾病风险比为2.83(95%CI 2.62-3.05)。

核心结论

  • MoChiAgent可提供临床相关、可操作的决策支持信息,以加强母婴风险分层护理。

研究设计与样本

  • 开发并内部评估MoChiFormer,使用4,401,599次纵向临床就诊数据。
  • 外部验证使用独立母婴队列,分别包含263,452和23,192次就诊。

主要发现

  • MoChiFormer准确识别关键妊娠疾病,胎盘早剥AUC为0.89,胎膜早破AUC为0.89,早产AUC为0.91。
  • 配对母婴数据分析显示跨代风险关联,特定母亲集群所生婴儿新生儿黄疸风险比为2.81(95%CI 2.60-3.03),血液疾病风险比为2.83(95%CI 2.62-3.05)。
  • 整合母亲妊娠期电子健康记录与婴儿记录可改善对婴儿疾病(包括染色体异常和呼吸系统疾病)的预测。

临床怎么用

  • MoChiAgent可利用常规实验室检查等纵向电子健康记录数据预测母婴疾病,提供循证干预和治疗建议。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

Current predictive models for pregnancy and infant outcomes often focus on limited endpoints and rely on costly tests or imaging. Here we developed the Mother-Child Artificial Intelligence Agent (MoChiAgent)-a large-language-model-based clinical assistant that orchestrates several tools to integrate sequential electronic health record (EHR) data, including routine laboratory tests, for forecasting maternal and infant diseases. MoChiAgent's core predictive engine, MoChiFormer, was developed and evaluated internally using 4,401,599 longitudinal clinical visits and validated externally using independent maternal and infant cohorts consisting of 263,452 and 23,192 visits, respectively. MoChiFormer reconstructs missing laboratory values, reduces batch effects and learns EHR representations that support gestational, fetal and infant age estimation, health-trajectory modeling and stratification of current and future disease risk. Subsequently, a Knowledge Search Tool utilizes these forecasts to retrieve evidence-based intervention and treatment recommendations from curated medical literature and authoritative guidelines. For maternal health, MoChiFormer accurately identified key gestational conditions, achieving areas under the receiver operating characteristic curves of 0.89 for placental abruption, 0.89 for premature rupture of membranes and 0.91 for preterm labor. Analysis of paired mother-infant data further revealed transgenerational risk associations, with infants born to mothers in specific clusters showing substantially elevated risks of neonatal jaundice (hazard ratio = 2.81; 95% confidence interval, 2.60-3.03) and hematological diseases (hazard ratio = 2.83; 95% confidence interval, 2.62-3.05). Integrating maternal gestational EHRs with infant records improved prediction of infant conditions, including chromosomal abnormalities and respiratory disorders. These findings indicate that MoChiAgent can provide clinically relevant, actionable decision support information to enhance risk-stratified care for mothers and infants. ClinicalTrials.gov identifier: NCT06791486 .

来源:Nat Med,查看原文。版权归原作者所有。

引用

  1. Liu S, Zheng W, Kang J, et al. Prediction of maternal and infant outcomes from longitudinal electronic health records with a mother-child AI agent. Nat Med. 2026. PMID: 42742183. doi:10.1038/s41591-026-04694-y

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。