External validation set Sample Clauses

External validation set. IEO Table 1. Performance of the ensemble methodology using the IEO data set for external validation. Model 1 Model 2 Model 3 Model 4

External validation set. HUJI In Table 3 the results of the mental health and QoL risk prediction models are given when the HUJI cohort was considered for validation purposes according to the different approaches (i.e., inclusion or not of anxiety and depression symptoms and QoL predictors at M0 and M3).

External validation set. CHAMP In Table 4 the results of the mental health and QoL risk prediction models are given when the CHAMP cohort was considered for validation purposes according to the different approaches. Accuracy 0.761 0.787 0.640 0.663 Balanced accuracy 0.801 0.834 0.739 0.687 AUC 0.879 0.913 0.807 0.783 Sensitivity 0.869 0.913 0.923 0.730 Specificity 0.733 0.755 0.556 0.643 F1 0.597 0.636 0.539 0.500 Note. Cross validation using grid search optimization against Area Under the ROC Curve (AUC) was applied following a 3-fold data division scheme for better and more generalizable results. The figure below ranks predictor variables entered into Model 1 by feature importances computed based on the mean decrease of the impurity metric.