使用常规生物标志物的可解释机器学习识别了培养定义的细菌性尿素
Yuan-Lu Zhang1, Dong-Xiao Yu2, Ying-Ying Zheng1
1Department of Clinical Laboratory, Fuding Hospital, Fujian University of Traditional Chinese Medicine, Fuding, 355200, Fujian Province, China.
Scientific reports
|March 4, 2026
概括
早期识别细菌性尿路,严重的尿路感染并发症,至关重要. 机器学习模型使用常规生物标记物,如D-二聚体和prokalcitonin,显示出在住院患者中风险分层的前景.
科学领域:
- 临床医学 临床医学
- 传染性疾病 传染性疾病
- 医疗信息学 医疗信息学
背景情况:
- 尿路炎症是一种严重的尿路感染 (UTI) 并发症,具有严重的器官功能障碍和死亡风险.
- 早期和准确地识别细菌性尿素仍然是一个临床挑战,需要改进风险分层工具.
- 常规可用的实验室数据为开发用于早期风险评估的预测模型提供了潜在的途径.
研究的目的:
- 开发和评估机器学习模型 (随机森林,XGBoost,物流回归) 用于细菌性尿素的早期风险分层.
- 为了评估模型的预测性能,使用常规生物标志物在患者呈现后24小时内获得.
- 在培养确认尿路感染的住院患者中,确定细菌性尿路的关键常规实验室预测因子.
主要方法:
- 来自 182 名住院患者的临床数据的回顾性分析,其中有培养确认的尿路感染.
- 开发随机森林 (RF),极端梯度提升 (XGBoost) 和多变量后勤回归 (LR) 模型,使用0-24h的生物标志物.
- 通过在持久测试集上的接收器运行特征曲线 (AUC) 下面的面积来评估模型歧视.
主要成果:
- 在XGBoost模型中,AUC达到0.886,在测试组中表现优于RF (0.822) 和LR (0.822),尽管差异在统计学上并不显著.
- 确定的主要预测因子包括D-二聚合物,前素 (PCT),C反应蛋白 (CRP),白细胞计数 (WBC) 和专蛋白.
- 与非细菌性尿路感染相比,细菌性尿路感染病例呈现显著更高的PCT,CRP,WBC和较低的白蛋白.
结论:
- 机器学习模型,特别是XGBoost,在24小时内使用常规生物标志物识别细菌性尿素,表现出良好的区分能力.
- D-二聚合物,前素和白蛋白是重要的预测因子,突出了常规实验室检测早期风险分层的潜力.
- 建议进行外部验证,以确认这些模型在各种临床环境中的实用性.
更多相关视频
相关概念视频
Urine Studies II: Urine Culture and Sensitivity Test
1.7K
A urine culture and sensitivity test is a diagnostic procedure used to identify urinary tract bacterial infections and determine the most effective antibiotics for treatment. This test is generally preferred when a patient shows manifestations of a urinary tract infection, such as frequent or painful urination, cloudy or foul-smelling urine, or lower abdominal pain.Purpose of the TestThe primary goals of a urine culture and sensitivity test are to:Determine the specific bacteria causing the...
1.7K
Methods of Classification and Identification
1.4K
Bacterial identification relies on a diverse array of techniques to classify and understand microorganisms, each tailored to uncover specific characteristics. Traditional morphological approaches, while still valuable, are limited for closely related or structurally simple organisms. Modern methods integrate biochemical, serological, genetic, and advanced molecular tools to achieve greater accuracy.Morphological and Biochemical TechniquesMorphological characteristics, such as cell shape and...
1.4K
Steps in Outbreak Investigation
656
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
656


