Table 2 Performance metrics of the DL models on the datasets.

From: Ensemble learning for fetal ultrasound and maternal–fetal data to predict mode of delivery after labor induction

Model

AUROC

F1-score

PPV

NPV

Sensitivity

Specificity

Tabular data models

 AdaBoost

0.738 ± 0.051

0.736 ± 0.024

0.734 ± 0.024

0.790 ± 0.019

0.746 ± 0.022

0.867 ± 0.019

 Logistic regression

0.772 ± 0.031

0.731 ± 0.039

0.728 ± 0.040

0.789 ± 0.027

0.740 ± 0.036

0.856 ± 0.030

 MLP

0.739 ± 0.024

0.730 ± 0.010

0.727 ± 0.011

0.794 ± 0.013

0.736 ± 0.007

0.840 ± 0.011

 RandomForest

0.769 ± 0.031

0.725 ± 0.027

0.730 ± 0.027

0.772 ± 0.020

0.746 ± 0.021

0.902 ± 0.016

 XGBoost

0.729 ± 0.039

0.705 ± 0.015

0.701 ± 0.016

0.775 ± 0.014

0.713 ± 0.011

0.829 ± 0.006

 SVC

0.732 ± 0.023

0.587 ± 0.003

0.727 ± 0.112

0.700 ± 0.001

0.697 ± 0.010

0.988 ± 0.021

Image data models

 Inception femur

0.485 ± 0.019

0.605 ± 0.015

0.596 ± 0.017

0.703 ± 0.006

0.668 ± 0.003

0.908 ± 0.022

 Inception head

0.492 ± 0.036

0.584 ± 0.028

0.570 ± 0.043

0.693 ± 0.013

0.655 ± 0.036

0.904 ± 0.052

 Inception abdomen

0.487 ± 0.034

0.584 ± 0.017

0.563 ± 0.025

0.692 ± 0.009

0.645 ± 0.015

0.883 ± 0.011

 Xception femur

0.505 ± 0.075

0.597 ± 0.035

0.573 ± 0.053

0.700 ± 0.016

0.653 ± 0.011

0.881 ± 0.039

 Xception Head

0.460 ± 0.038

0.578 ± 0.017

0.559 ± 0.017

0.689 ± 0.010

0.629 ± 0.039

0.849 ± 0.082

 Xception Abdomen

0.514 ± 0.015

0.589 ± 0.019

0.576 ± 0.023

0.698 ± 0.006

0.673 ± 0.011

0.904 ± 0.032

 ResNet50 femur

0.492 ± 0.077

0.575 ± 0.007

0.514 ± 0.052

0.697 ± 0.002

0.693 ± 0.005

0.991 ± 0.015

 ResNet50 head

0.548 ± 0.046

0.586 ± 0.014

0.680 ± 0.103

0.700 ± 0.005

0.698 ± 0.006

0.991 ± 0.008

 ResNet50 abdomen

0.513 ± 0.062

0.571 ± 0.002

0.485 ± 0.003

0.696 ± 0.002

0.694 ± 0.003

0.996 ± 0.003

  1. Results are presented as mean ± standard deviation. These values were obtained by threefold cross-validation. Columns include: AUROC: area under the receiver operating curve; PPV: positive predictive value; NPV: negative predictive value.
Back to article page