剖宫产术后低体温的机器学习预测模型构建

朱 玉, 梁 仁瑞
湛江中心人民医院

摘要


目的：本研究通过构建并验证剖宫产术后低体温的预测模型，并筛选关键风险因素，为临床护理早期干预提供依据。方法：本研究回顾性分析剖宫产患者的临床资料，并通过采用多种机器学习算法构建预测术后剖宫产低体温模型，通过训练组和测试组验证模型性能，并筛选影响术后低体温的关键影响变量。结果：XGBoost模型在训练集（准确率0.976）和测试集（准确率0.890）表现最优；Random Forest算法筛选出的前6个关键风险因素为BMI、体重、手术室温度、术前脚趾灌注指数、身高和白蛋白。结论：基于机器学习的剖宫产术后低体温预测模型可有效识别术后剖宫产低体温高危人群，为临床护理精准干预提供新策略。

关键词


剖宫产；术后低体温；预测模型；机器学习；风险因素

全文:

PDF


参考


[1]Mei A, Gao L, Wang Y, et al. Establishment and Validation of Prediction Model for Post-Operative Hypothermia During the Post-Anaesthesia Care Unit Stay. J Clin Nurs. 2025;34(7):3017-3026. doi:10.1111/jocn.17815

[2]Dandan Z, Yiting W, Shengqiang Z, Tiantian Z, Jing Y. Systematic review of prediction models for post-traumatic hypothermia risk. Injury. 2025;56(12):112883. doi:10.1016/j.injury.2025.112883

[3]Cao B, Li Y, Chen X, et al. Development and validation of a novel risk assessment model for accurate prediction of intraoperative hypothermia in adult patients undergoing different types of surgery: insights from a multicentre, retrospective cohort study. Ann Med. 2025;57(1):2489749. doi:10.1080/07853890.2025.2489749

[4]Jiang J, Feng S, Sun Y, An Y. Risk factors for hypothermia after transurethral holmium laser enucleation of the prostate and development of a nomogram model. 经尿道钬激光前列腺剜除术后低体温的危险因素及列线图预测模型的构建. Zhong Nan Da Xue Xue Bao Yi Xue Ban. 2024;49(11):1741-1750. doi:10.11817/j.issn.1672-7347.2024.240460

[5]Hu Z, Li W, Liang C, Li K. Risk factors and prediction model for inadvertent intraoperative hypothermia in patients undergoing robotic surgery: a retrospective analysis. Sci Rep. 2023;13(1):3687. Published 2023 Mar 6. doi:10.1038/s41598-023-30819-1


Refbacks

  • 当前没有refback。