基于机器学习预测妊娠期糖尿病母亲分娩大于胎龄儿
本研究基于371例妊娠期糖尿病(GDM)女性的前瞻性队列,利用20至34孕周的连续血糖监测(CGM)数据、人口学数据和代谢指标,开发并验证了一种基于卷积神经网络(CNN)的融合模型来预测大于胎龄儿(LGA)。371例中76例(20.5%)分娩LGA新生儿;融合模型预测LGA的AUCROC为0.852(95% CI,0.680-0.966),准确率84.4%,判别能力优于3种传统模型。
要点
- 371例GDM女性中76例(20.5%)分娩LGA新生儿。
- 基于24小时CGM数据和15项临床变量的AI融合模型预测LGA的AUCROC为0.852(95% CI,0.680-0.966),准确率84.4%。
- 融合模型判别能力优于3种传统模型,CGM图谱特征有助于确定干预窗口。
核心结论
- 基于AI的融合模型在预测GDM女性分娩LGA婴儿方面表现出更好的性能。
- CGM图谱特征有助于确定适当的干预窗口。
研究设计与样本
- 前瞻性队列研究,纳入大学医院371例GDM女性。
- CGM在20至34孕周进行,评估并可视化分娩LGA与非LGA婴儿的GDM女性的血糖波动。
- 开发基于卷积神经网络(CNN)的融合模型预测LGA,并与3种传统模型比较,使用AUCROC和准确率评价。
主要发现
- 371例GDM女性中76例(20.5%)分娩LGA新生儿。
- 可视化24小时血糖谱在上午中间时段存在差异,该差异在按孕前体重指数、治疗方案和CGM实施时期分层的亚组中一致。
- 使用24小时CGM数据和15项临床变量的AI融合预测模型预测LGA的AUCROC为0.852(95% CI,0.680-0.966),准确率84.4%,判别能力优于3种传统模型。
局限性
- 原文未报告局限性。
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
CONTEXT: Large-for-gestational-age (LGA), one of the most common complications of gestational diabetes mellitus (GDM), has become a global concern. The predictive performance of common continuous glucose monitoring (CGM) metrics for LGA is limited. OBJECTIVE: We aimed to develop and validate an artificial intelligence (AI)-based model to determine the probability of women with GDM giving birth to LGA infants during pregnancy using CGM measurements together with demographic data and metabolic indicators. METHODS: A total of 371 women with GDM from a prospective cohort at a university hospital were included. CGM was performed during 20 to 34 gestational weeks, and glycemic fluctuations were evaluated and visualized in women with GDM who gave birth to LGA and non-LGA infants. A convolutional neural network (CNN)-based fusion model was developed to predict LGA. Comparisons among the novel fusion model and 3 conventional models were made using the area under the receiver operating characteristic curve (AUCROC) and accuracy. RESULTS: Overall, 76 (20.5%) out of 371 GDM women developed LGA neonates. The visualized 24-hour glucose profiles differed at midmorning. This difference was consistent among subgroups categorized by pregestational body mass index, therapeutic protocol, and CGM administration period. The AI-based fusion prediction model using 24-hour CGM data and 15 clinical variables for LGA prediction (AUCROC 0.852; 95% CI, 0.680-0.966; accuracy 84.4%) showed superior discriminative power compared with the 3 classic models. CONCLUSION: We demonstrated better performance in predicting LGA infants among women with GDM using the AI-based fusion model. The characteristics of the CGM profiles allowed us to determine the appropriate window for intervention.
引用
- Kang M, Zhu C, Lai M, et al. Machine Learning-Based Prediction of Large-for-Gestational-Age Infants in Mothers With Gestational Diabetes Mellitus. J Clin Endocrinol Metab. 2025. PMID: 39011974. doi:10.1210/clinem/dgae475
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。