Fig. 3From: Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scanValidation AUC for each pair of dataset and machine learning algorithm, when lung recruitability was gas exchange-defined (recruiters: Δ15-5PaO2 > 24 mmHg). M5, lung mechanics at PEEP 5 cmH2O, M15, lung mechanics at PEEP 15 cmH2O, RPM, respiratory partitioned mechanics, G5, gas exchange measured at PEEP 5 cmH2O, CT5, CT imaging acquired at PEEP 5 cmH2O. XGBoost, gradient-boosted tree; RF, random forest; LR, logistic regression; SVM, support vector machineBack to article page