Table 2 Cross-validated discriminative performance of the Super Learner algorithm for predicting small for gestational age and large for gestational age fitted using grandmaternal pregnancy-related information and maternal birth characteristics (G0 predictors), maternal clinical factors at 26 weeks’ gestation (G1 predictors), and their combination (G0 + G1 predictors).

From: Development and validation of super learner models to predict small and large for gestational age in the second generation

 

AUC-ROC (95% CI)

AUC-PR (95% CI)

Mean Brier Scorea

Small for gestational age

G0 predictors

0.63 (0.61, 0.65)

0.15 (0.14, 0.17)

0.0877

G1 predictors

0.66 (0.64, 0.68)

0.18 (0.16, 0.21)

0.0864

G0 + G1 predictors

0.69 (0.67, 0.71)

0.21 (0.19, 0.24)

0.0850

Large for gestational age

G0 predictors

0.64 (0.62, 0.67)

0.17 (0.15, 0.18)

0.0862

G1 predictors

0.66 (0.64, 0.68)

0.18 (0.16, 0.20)

0.0857

G0 + G1 predictors

0.71 (0.69, 0.73)

0.22 (0.19, 0.24)

0.0837

  1. AUC-PR area under the precision-recall curve; AUC-ROC area under the receiver operating characteristic curve; CI confidence interval.
  2. a Mean Brier score across validation folds and imputed data sets.
Back to article page