机器学习预测模型用于ICU呼吸器相关肺炎患者的多药耐药生物感染:使用MIMIC-IV数据库进行分析
Zhigang Cui1, Yifan Dong2, Huizhu Yang3
1School of Nursing, China Medical University, Shenyang, Liaoning, China.
Computers in biology and medicine
|March 28, 2025
概括
机器学习模型可以预测呼吸机相关性肺炎 (VAP) 患者的多药耐药生物体 (MDRO) 感染. 在VAP中,XGBoost在识别MDRO的高风险因素方面表现出卓越的表现.
科学领域:
- 医疗信息学 医疗信息学
- 计算生物学 计算生物学
- 传染性疾病 传染性疾病
背景情况:
- 呼吸机相关肺炎 (VAP) 是重症监护病房中死亡的一个重要原因.
- 多药耐药生物体 (MDRO) 感染复杂化VAP,增加发病率和死亡率.
- 在VAP中对MDROs的现有预测模型可能有局限性,需要改进方法.
研究的目的:
- 开发和比较四种机器学习模型,用于识别VAP患者中MDRO感染的高风险因素.
- 利用MIMIC-IV数据库构建可靠的预测模型.
- 评估这些模型的临床实用性和可解释性.
主要方法:
- 使用了MIMIC-IV数据库,包括972名VAP患者.
- 应用合成少数群体过量采样技术 (SMOTE) 对于类不平衡.
- 采用LASSO回归和特征选择的特征重要性.
- 构建并比较了后勤回归,XGBoost,随机森林和梯度增强机器模型.
- 使用ROC曲线,校准曲线,Brier分数和决策曲线分析 (DCA) 评估模型性能.
- 使用SHAP值的解释模型.
主要成果:
- 在测试组中,XGBoost获得了最高的预测性能,AUC为0.831 (95% CI:0.785-0.877).
- 确定了VAP中MDRO的20个关键风险因素,包括红细胞分布宽度,机械通风的持续时间,离子间隙,基和中性百分比.
- 模型通过DCA证明了良好的校准和临床实用性.
- 在MDROs-VAP和非MDROs-VAP组之间观察到显著的临床参数差异.
结论:
- XGBoost成为预测VAP患者MDRO感染的最佳机器学习模型.
- SHAP分析为20个独立的风险因素提供了宝贵的见解,证实了该模型的预测能力.
- 该研究强调了机器学习在增强VAP中MDROs预测方面的潜力,需要通过可靠的数据和方法进一步验证.
相关概念视频
01:28Pneumonia IV: Management
The treatment of pneumonia varies based on its severity and the causative pathogen. Here is a structured approach to managing pneumonia, integrating pharmaceutical and supportive care strategies.
Bacterial Pneumonia Treatment
For bacterial pneumonia, antibiotics serve as the cornerstone of therapy. Initial treatment often begins with empirical antibiotics, tailored to the anticipated causative organism and adjusted based on culture results. Key antibiotic choices include:
Bacterial Pneumonia Treatment
For bacterial pneumonia, antibiotics serve as the cornerstone of therapy. Initial treatment often begins with empirical antibiotics, tailored to the anticipated causative organism and adjusted based on culture results. Key antibiotic choices include:
01:21Acute Respiratory Failure-II
Type I Respiratory Failure, or hypoxemic respiratory failure, occurs when the partial pressure of oxygen (PaO2) in arterial blood falls below 60 mmHg while breathing room air without a corresponding increase in arterial carbon dioxide levels (PaCO2). This condition highlights a significant impairment in the lungs' capacity to oxygenate the blood.
The underlying physiological abnormalities that contribute to hypoxemic respiratory failure include:
The underlying physiological abnormalities that contribute to hypoxemic respiratory failure include:
01:23Mechanical Ventilation II: Invasive Ventilation
Ventilators are essential medical equipment used to aid patients with respiratory difficulties. Their primary function is to assist or replace spontaneous breathing by providing mechanical ventilation. There are two general classes of mechanical ventilators: negative-pressure and positive-pressure ventilators.
Negative-Pressure Ventilators
Negative-pressure ventilators create a vacuum around the chest or body to draw air into the lungs, simulating breathing. This method does not require an...
Negative-Pressure Ventilators
Negative-pressure ventilators create a vacuum around the chest or body to draw air into the lungs, simulating breathing. This method does not require an...