What question did this study set out to answer?

The aim is to develop a machine learning tool to predict acute respiratory distress syndrome in polytraumatic patients.

March 26, 2026Open Access

An Ensemble Learning Approach for Predicting Acute Respiratory Distress Syndrome Within Polytraumatic Patients In Imbalanced Dataset

Key Points

The aim is to develop a machine learning tool to predict acute respiratory distress syndrome in polytraumatic patients.
Retrospective analysis of data from 407 polytrauma patients
Collection of patient data at admission and 12 to 24 hours later
Application of Random Forest classifier to identify key predictors
Evaluation using metrics like AUROC, PR-AUC, F1 score, recall, and precision
Utilization of SHAP values to interpret model predictions
Model achieved an AUROC of 0.96 indicating high predictive accuracy
PR-AUC was 0.81, showing good performance in classifying incidents
Recall of 0.92 suggests effective identification of true ARDS cases
F1 score of 0.76 indicates a balance between precision and recall
Identified key predictors include arterial oxygen pressure, flail chest injury, blood pH, hemothorax, and fluid resuscitation

Abstract

Acute Respiratory Distress Syndrome (ARDS) is a severe, often overlooked complication in polytrauma patients, complicated by its varied symptoms and delayed diagnosis through imaging. To address this, our study developed a machine learning tool to predict ARDS early, using data from 407 polytrauma patients collected at admission, 12 hours, and 24 hours later. We aimed to create a clear, practical decision-support tool for clinicians. For this, we retrospectively analyzed de-identified patient data from a clinical observational group. To improve predictive accuracy, we handled data imbalance and selected the most informative clinical features. A Random Forest classifier was trained and evaluated with metrics like AUROC, PR-AUC, F1 score, recall, and precision, while SHAP values helped explain the model’s predictions. The model performed strongly, achieving an AUROC of 0.96, PR-AUC of 0.81, recall of 0.92, and F1 score of 0.76. Key predictors included arterial oxygen pressure (PaO2), flail chest injury, blood pH at 24 hours, hemothorax, and total fluid resuscitation, offering clinicians actionable insights for early ARDS detection. This ML model permits early prediction of ARDS in polytraumatic patients, providing potential for eventual incorporation into electronic health records to minimize mortality through in-time interventions.

Demander à l'IA

Bookmark

View Full Paper