You are viewing the site in preview mode

Skip to main content

Advertisement

Table 3 Area under the curve, accuracy, and the three most important predictors for the prediction of large for gestational age (LGA) birth using logistic regression and five machine learning methods pre-pregnancy and at 26 weeks in nulliparous and multiparous women

From: Comparison of logistic regression with machine learning methods for the prediction of fetal growth abnormalities: a retrospective cohort study

  Pre-pregnancy 26 weeks
LR EN CT RF GB NN LR EN CART RF GB NN
LGA - Primiparae
 Area under the curve 0.592 0.587 0.563 0.576 0.587 0.594 0.702 0.705 0.675 0.673 0.697 0.705
 Accuracy 0.826 0.827 0.800 0.824 0.832 0.827 0.843 0.834 0.780 0.834 0.839 0.842
 Maternal age          
 Common-law/married            
 Pre-pregnancy smoking          
 Pre-pregnancy BMI    
 Pre-existing diabetes          
 Weight gain at 26 wks         
 Smoking in pregnancy           
 Pregnancy-induced hypertension            
 Gestational diabetes             
LGA - Multiparae
 Area under the curve 0.700 0.700 0.659 0.692 0.704 0.700 0.745 0.748 0.718 0.728 0.748 0.746
 Accuracy 0.807 0.806 0.817 0.795 0.804 0.807 0.813 0.809 0.794 0.799 0.805 0.812
 Maternal age            
 Pre-pregnancy smoking             
 Pre-pregnancy BMI       
 Pre-existing diabetes            
 Previous LBW infant             
 Previous infant > 4080 g
 Previous death of neonate ≥500 g            
 Weight gain at 26 wks         
 Smoking in pregnancy             
  1. Abbreviations: BMI body mass index, CT classification tree, EN elastic net, GB Gradient boosting, LBW low birth weight, LR logistic regression, NN neural network, RF random forest, wks weeks