在关键性IBD护理中随机森林驱动的死亡率预测:结合并发病模式和实时生理学的双数据库模型
Zhenze Zhang1,2, Caiqing Zhao1,3, Yijun Zhou4
1Clinical Graduate School, Jiangxi Medical College, Nanchang University, Nanchang, China.
Frontiers in medicine
|August 27, 2025
概括
机器学习模型可以准确预测炎症性肠病 (IBD) 严重患者的死亡率. 这项研究采用了使用ICU数据的验证名录,以改善这些高风险患者的风险分层.
科学领域:
- 危急护理医学
- 胃肠病学
- 数据科学与机器学习
背景情况:
- 在重症监护室 (ICU) 住院的炎症性肠病 (IBD) 患者因并发症而面临严重的死亡风险.
- 目前用于预测严重IBD患者的死亡率的预测工具有限.
- 通过分析复杂的临床数据,机器学习有可能改善风险分层.
研究的目的:
- 开发和验证基于ML的模型,用于重症IBD患者的死亡风险分层.
- 确定该患者队列中死亡率的关键预测因素.
- 创建一个可解读的临床用品.
主要方法:
- 来自MIMIC- IV数据库的551名IBD患者的分析,使用eICU数据集进行外部验证.
- 训练和评估9个ML算法来预测1年的死亡率.
- 使用人口统计学,并发症,实验室结果,生命体征和严重性得分作为预测因素,以SHAP为特征重要性和物流回归为名ogram构建.
主要成果:
- 随机森林模型在内部验证中显示出优异的区别 (AUC> 0.8).
- 确定的关键预测因素包括恶性病史,查尔森并发症指数 (CCI),红细胞分布宽度 (RDW),格拉斯哥昏迷表 (GCS),序列器官衰竭评估 (SOFA),年龄,心率,体重和性别.
- 在eICU队列中,开发的名图显示出强大的外部验证性能 (AUC> 0. 8).
结论:
- 一种基于ML的新型名谱被开发和验证,用于预测重症IBD患者的死亡率.
- 该名录集成了从多中心ICU数据中获得的可解释预测因素.
- 这种工具可以帮助对这一脆弱群体进行风险分层和临床决策.
相关概念视频
Rheumatic Heart Disease IV: Nursing Management
37
AssessmentA comprehensive assessment is essential in managing a patient with rheumatic heart disease (RHD). Begin with obtaining a detailed medical history, including recent streptococcal infections, a history of rheumatic fever, or previously diagnosed rheumatic heart disease. Assess the patient for symptoms such as fever, chest pain, widespread joint pain (arthralgia), tachycardia, pericardial friction rub, muffled heart sounds, heart murmurs, peripheral edema, subcutaneous nodules, and...
37
Model Approaches for Pharmacokinetic Data: Physiological Models
109
Physiological models in pharmacokinetics are instrumental in understanding the distribution and elimination of drugs within the body. These models describe the drug concentration within target organs, influenced by factors such as drug uptake, tissue volume, and blood flow. Drug uptake is governed by the partition coefficient, which signifies the drug concentration ratio in tissue to that in the blood. The blood flow rate to a specific tissue is expressed as Qt, and the rate of change in tissue...
109


