无监督机器学习分析,以确定ICU药物使用模式,以预测液体过载
Kelli Henry1, Shiyuan Deng2, Xianyan Chen2
1Department of Pharmacy, Wellstar MCG Health, Augusta, Georgia, USA.
机器学习确定了一个与重症监护室 (ICU) 液体过载 (FO) 密切相关的药物集群. 这一发现改善了FO的预测,可能使早期干预成为更好的患者结果.
科学领域:
- 密集护理医学是密集护理的医学.
- 在医疗保健中的数据科学.
- 药理学 药理学是指药理学的学科.
背景情况:
- 液体过载 (FO) 是重症监护室 (ICU) 中常见且严重的并发症.
- 静脉注射药物是FO的主要原因之一,但其复杂的药物管理模式使其难以预测.
- 无监督机器学习可以识别与FO相关的药物管理模式.
研究的目的:
- 应用无监督机器学习来发现与FO相关的药物管理模式.
- 评估已识别的药物模式对FO发展的预测价值.
主要方法:
- 对927名成年ICU患者进行回顾性队列研究,停留时间≥72小时.
- FO定义为体液平衡≥7%的入院体重.
- 使用主要成分分析 (PCA) 和受限制的博尔茨曼机器 (RBM) 在3小时间隔分析药物管理数据,以确定药物集群.
主要成果:
- 流体过载 (FO) 在13.7%的患者中发生.
- 从47,803次静脉注射药物管理中确定了10个独特的药物集群.
- 一个特定的药物集群 (集群7),包括连续输液,抗生素和镇静剂/止痛药,与FO显著相关 (平均给药量:25.6在FO中vs10.9没有FO).
- 将Cluster 7药物添加到现有的预测模型中,改善了FO预测的准确性 (AUROC从0.719降至0.741).
结论:
- 无监督机器学习发现了一种新型药物集群,与ICU中的FO密切相关.
- 与传统模型相比,这种药物集群显著改善了FO预测.
- 将这种方法整合到临床实践中可能会增强早期FO检测和及时干预.
更多相关视频
07:59Bedside Ultrasound for Guiding Fluid Removal in Patients with Pulmonary Edema: The Reverse-FALLS Protocol
Published on: July 28, 2018
10:38Observational Study Protocol for Repeated Clinical Examination and Critical Care Ultrasonography Within the Simple Intensive Care Studies
Published on: January 16, 2019
相关概念视频
Heart Failure Drugs: Diuretics
One-Compartment Model: IV Infusion
The one-compartment model for IV infusion uses mathematical equations to describe the rate of change in drug quantity in the body. At steady-state or infusion equilibrium, the drug input...
One-Compartment Open Model for IV Bolus Administration: Estimation of Clearance
In the one-compartment open model for intravenous (IV) bolus administration, clearance is estimated by dividing the elimination rate by the plasma drug concentration. This equation leverages the elimination rate constant and the apparent...
Two-Compartment Open Model: IV Infusion
The model illustrates the decrease in plasma drug concentration from the central compartment with a specific equation. It shows that under steady-state conditions, the drug's input rate...
One-Compartment Open Model for IV Bolus Administration: General Considerations
The drug's presence in the body is defined by an equation representing the difference between the rates of drug entry and exit. Key parameters—elimination rate constant,...
Model-Independent Approaches for Pharmacokinetic Data: Noncompartmental Analysis
One important characteristic of noncompartmental analyses is that drug exposure increases proportionally with increasing doses. This...
