应用 XGBoost 机器学习方法预测糖尿病酮症酸中毒相关急性肾损伤的危险因素。

Predicting the risk factors of diabetic ketoacidosis-associated acute kidney injury: A machine learning approach using XGBoost.

机构信息

Department of Endocrinology, Second Affiliated Hospital of Jilin University, Changchun, China.

Digestive Diseases Center, Department of Hepatopancreatobiliary Medicine, Second Affiliated Hospital of Jilin University, Changchun, China.

出版信息

Front Public Health. 2023 Apr 6;11:1087297. doi: 10.3389/fpubh.2023.1087297. eCollection 2023.

DOI:10.3389/fpubh.2023.1087297

PMID:37089510

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC10117643/

Abstract

OBJECTIVE

The purpose of this study was to develop and validate a predictive model based on a machine learning (ML) approach to identify patients with DKA at increased risk of AKI within 1 week of hospitalization in the intensive care unit (ICU).

METHODS

Patients diagnosed with DKA from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database according to the International Classification of Diseases (ICD)-9/10 code were included. The patient's medical history is extracted, along with data on their demographics, vital signs, clinical characteristics, laboratory results, and therapeutic measures. The best-performing model is chosen by contrasting the 8 Ml models. The area under the receiver operating characteristic curve (AUC), sensitivity, accuracy, and specificity were calculated to select the best-performing ML model.

RESULTS

The final study enrolled 1,322 patients with DKA in total, randomly split into training (1,124, 85%) and validation sets (198, 15%). 497 (37.5%) of them experienced AKI within a week of being admitted to the ICU. The eXtreme Gradient Boosting (XGBoost) model performed best of the 8 Ml models, and the AUC of the training and validation sets were 0.835 and 0.800, respectively. According to the , the top 5 main features contributing to the XGBoost model were blood urea nitrogen (BUN), urine output, weight, age, and platelet count (PLT).

CONCLUSION

An ML-based individual prediction model for DKA-associated AKI (DKA-AKI) was developed and validated. The model performs robustly, identifies high-risk patients early, can assist in clinical decision-making, and can improve the prognosis of DKA patients to some extent.

摘要

目的

本研究旨在开发和验证一种基于机器学习（ML）方法的预测模型，以识别在重症监护病房（ICU）住院 1 周内发生 AKI 的 DKA 患者的风险增加。

方法

根据国际疾病分类（ICD）第 9/10 版代码，从医疗信息集市重症监护 IV（MIMIC-IV）数据库中纳入 DKA 患者。提取患者的病史，以及人口统计学、生命体征、临床特征、实验室结果和治疗措施的数据。通过对比 8 个 ML 模型选择性能最佳的模型。计算接收者操作特征曲线（AUC）、灵敏度、准确性和特异性，以选择性能最佳的 ML 模型。

结果

最终研究共纳入 1322 例 DKA 患者，随机分为训练集（1124 例，85%）和验证集（198 例，15%）。其中 497 例（37.5%）在 ICU 住院 1 周内发生 AKI。在 8 个 ML 模型中，极端梯度提升（XGBoost）模型的性能最佳，训练集和验证集的 AUC 分别为 0.835 和 0.800。根据 XGBoost 模型，对模型贡献最大的前 5 个主要特征是血尿素氮（BUN）、尿量、体重、年龄和血小板计数（PLT）。