ARDSMLpred: A machine learning model for predicting SCAP-associated ARDS based on MIMIC-IV database

BACKGROUND: Early identification of acute respiratory distress syndrome (ARDS) in severe community-acquired pneumonia (SCAP) are crucial for reducing morbidity and mortality. This study focuses on developing an optimal prediction model based on clinical data and biomarkers to detect the risk of SCAP-associated ARDS in adult ICU patients. METHODS: This investigation, utilizing the MIMIC-IV database, enrolled 3,807 patients with SCAP and randomly allocated them into a training set (n=2,664) for model development and a testing set (n=1,143) as an internal validation cohort to assess the model's predictive performance. The outcome was defined as the incidence of SCAP-associated ARDS. Baseline clinical and laboratory characteristics of the patients were obtained. Selection of characteristic variables was performed using LASSO regression, followed by the construction of ten ML models: LGBM, KNN, CatBoost, SVM, XGBoost, DesicionTree, NB, RF, KNNC, and MLP. The evaluation of model performance is conducted through various indicators such as ROC curves, calibration curve, DCA, accuracy, specificity, recall, presicion and F1 score. RESULTS: , is now available on our website (https://zxhzxh2000.shinyapps.io/ards/). CONCLUSION: The prediction model constructed based on 8 characteristic variables selected by LASSO regression and using the XGBoost algorithm has excellent predictive performance in predicting the occurrence of SCAP-associated ARDS in adult ICU patients. This data-driven predictive model will help clinicians to make quick and accurate diagnosis.