利用多种机器学习算法开发新生儿医院获得性胃肠道感染预测模型
Development of a Predictive Model for Neonatal Hospital-Acquired Gastrointestinal Infections Utilizing Multiple Machine Learning Algorithms.
作者信息
Shao Hui, Chen Huajuan, Wang Xiujuan
机构信息
Department of Infectology, Shaoxing Maternity and Child Health Care Hospital (Maternity and Child Health Care Affiliated Hospital, Shaoxing University), Shaoxing, People's Republic of China.
Department of Obstetrics, Shaoxing Maternity and Child Health Care Hospital (Maternity and Child Health Care Affiliated Hospital, Shaoxing University), Shaoxing, People's Republic of China.
出版信息
Infect Drug Resist. 2025 Aug 18;18:4141-4156. doi: 10.2147/IDR.S533904. eCollection 2025.
OBJECTIVE
To develop and validate a predictive model for neonatal gastrointestinal infections using multiple machine learning algorithms.
METHODS
We conducted a retrospective analysis of 176 neonates diagnosed with nosocomial gastrointestinal infections in NICU between 2020 and 2024, along with a randomly selected control group of 675 neonates without such diagnoses during their NICU stay. The study examined 29 perinatal and NICU treatment-related risk factors potentially associated with neonatal gastrointestinal infections. The dataset was randomly partitioned into training and testing sets. To address class imbalance and enhance minority class identification, we applied SMOTE to the training set.Feature selection used Boruta, Lasso, and Logistic regression, with consensus features from Venn analysis.Subsequently, eight machine learning algorithms were implemented to construct predictive models. Models were evaluated using AUC, F1 score, accuracy, sensitivity, and specificity.
RESULTS
The model incorporated nine significant feature variables: gestational age, NE, PLT, central venous catheterization, nasogastric feeding, delivery mode, intrauterine distress, pregnancy-induced hypertension, and probiotic administration. Among the eight machine learning algorithms evaluated, the Neural Network model demonstrated optimal performance - achieving perfect metrics in the training set (AUC=0.895, F1=0.845, Accuracy=0.837, Sensitivity=0.888, Specificity=0.786, Precision=0.806) and robust results in the test set (AUC= 0.876, F1=0.862, Accuracy=0.856, Sensitivity=0.896, Specificity=0.817, Precision=0.830) - thus was selected as the final predictive model. Model interpretability was enhanced through SHAP analysis. Furthermore, a Shiny-based interactive web calculator for neonatal gastrointestinal infection risk prediction was successfully developed based on this model.
CONCLUSION
The model effectively identifies at-risk neonates early, supporting clinical decision-making and timely interventions.
目的
使用多种机器学习算法开发并验证一种用于新生儿胃肠道感染的预测模型。
方法
我们对2020年至2024年期间在新生儿重症监护病房（NICU）被诊断为医院获得性胃肠道感染的176例新生儿进行了回顾性分析，并随机选取了675例在NICU住院期间未患此类疾病的新生儿作为对照组。该研究考察了29个与围产期和NICU治疗相关的潜在危险因素，这些因素可能与新生儿胃肠道感染有关。数据集被随机划分为训练集和测试集。为了解决类别不平衡问题并增强对少数类别的识别，我们对训练集应用了合成少数过采样技术（SMOTE）。特征选择使用了Boruta、套索回归和逻辑回归，并通过维恩分析得出共识特征。随后，实施了八种机器学习算法来构建预测模型。使用曲线下面积（AUC）、F1分数、准确率、敏感性和特异性对模型进行评估。
结果
该模型纳入了九个重要特征变量：胎龄、中性粒细胞计数、血小板计数、中心静脉置管、鼻饲、分娩方式、宫内窘迫、妊娠期高血压和益生菌使用。在评估的八种机器学习算法中，神经网络模型表现最优——在训练集中达到了完美的指标（AUC = 0.895，F1 = 0.845，准确率 = 0.837，敏感性 = 0.888，特异性 = 0.786，精确率 = 0.806），在测试集中也取得了稳健的结果（AUC = 0.876，F1 = 0.862，准确率 = 0.856，敏感性 = 0.896，特异性 = 0.817，精确率 = 0.830）——因此被选为最终的预测模型。通过SHAP分析增强了模型的可解释性。此外，基于该模型成功开发了一个基于Shiny的用于新生儿胃肠道感染风险预测的交互式网络计算器。
结论
该模型有效地早期识别出有风险的新生儿，支持临床决策和及时干预。
Zotero 插件