Internal Medicine Department, Staten Island University Hospital, Staten Island, NY, USA.
Department of Engineering, University of Massachusetts, Dartmouth, MA, USA.
Lupus. 2023 Oct;32(12):1418-1429. doi: 10.1177/09612033231206830. Epub 2023 Oct 13.
Although rare, severe systemic lupus erythematosus (SLE) flares requiring hospitalization account for most of the direct costs of SLE care. New machine learning (ML) methods may optimize lupus care by predicting which patients will have a prolonged hospital length of stay (LOS). Our study uses a machine learning approach to predict the LOS in patients admitted for lupus flares and assesses which features prolong LOS.
Our study sampled 5831 patients admitted for lupus flares from the National Inpatient Sample Database 2016-2018 and collected 90 demographics and comorbidity features. Four machine learning (ML) models were built (XGBoost, Linear Support Vector Machines, K Nearest Neighbors, and Logistic Regression) to predict LOS, and their performance was evaluated using multiple metrics, including accuracy, receiver operator area under the curve (ROC-AUC), precision-recall area under the curve (PR- AUC), and F1-score. Using the highest-performing model (XGBoost), we assessed the feature importance of our input features using Shapley value explanations (SHAP) to rank their impact on LOS.
Our XGB model performed the best with a ROC-AUC of 0.87, PR-AUC of 0.61, an F1 score of 0.56, and an accuracy of 95%. The features with the most significant impact on the model were "the need for a central line," "acute dialysis," and "acute renal failure." Other top features include those related to renal and infectious comorbidities.
Our results were consistent with the established literature and showed promise in ML over traditional methods of predictive analyses, even with rare rheumatic events such as lupus flare hospitalizations.
尽管罕见，但需要住院治疗的严重系统性红斑狼疮（SLE）发作占 SLE 治疗费用的大部分。新的机器学习（ML）方法可以通过预测哪些患者会有较长的住院时间（LOS）来优化狼疮护理。我们的研究使用机器学习方法来预测狼疮发作患者的 LOS，并评估哪些特征会延长 LOS。
我们的研究从 2016 年至 2018 年的国家住院患者样本数据库中抽取了 5831 例因狼疮发作入院的患者，并收集了 90 项人口统计学和合并症特征。我们构建了四个机器学习（ML）模型（XGBoost、线性支持向量机、K 最近邻和逻辑回归）来预测 LOS，并使用多种指标评估它们的性能，包括准确性、接收者操作特征曲线下的面积（ROC-AUC）、精度-召回曲线下的面积（PR-AUC）和 F1 分数。使用性能最高的模型（XGBoost），我们使用 Shapley 值解释（SHAP）评估输入特征的重要性，对 LOS 的影响进行排名。
我们的 XGB 模型表现最佳，ROC-AUC 为 0.87，PR-AUC 为 0.61，F1 得分为 0.56，准确率为 95%。对模型影响最大的特征是“需要中央静脉置管”、“急性透析”和“急性肾衰竭”。其他顶级特征包括与肾脏和感染合并症相关的特征。
我们的结果与已建立的文献一致，并表明在 ML 方面优于传统的预测分析方法，即使是在狼疮发作等罕见的风湿性事件住院治疗方面。
