Table 5 AUC, specificity, sensitivity, accuracy, NPV, and PPV of different models in the training and test groups.

From: Preoperative MRI-based radiomics analysis of intra- and peritumoral regions for predicting CD3 expression in early cervical cancer

Model

Group

AUC (95% CI)

Accuracy (95% CI)

Sensitivity (95% CI)

Specificity (95% CI)

PPV (95% CI)

NPV (95% CI)

mpMRI (ROItumor + ROI3mm)

Training group

0.906 (0.857–0.954)

0.823 (0.759–0.887)

0.823 (0.722–0.914)

0.823 (0.738, 0.904)

0.785 (0.677–0.881)

0.855 (0.770–0.929)

Test group

0.837 (0.718–0.956)

0.754 (0.639–0.853)

0.905 (0.750–1.000)

0.675 (0.531–0.816)

0.594 (0.424–0.765)

0.931 (0.828–1.000)

Clinical

Training group

0.738 (0.652–0.823)

0.731 (0.66–0.794)

0.776 (0.662–0.883)

0.699 (0.607–0.788)

0.643 (0.531–0.747)

0.817 (0.723–0.908)

Test group

0.703 (0.559–0.847)

0.754 (0.639–0.853)

0.840 (0.667–0.963)

0.694 (0.533–0.838)

0.656 (0.471–0.818)

0.862 (0.720–0.969)

Comprehensive

Training group

0.93 (0.88–0.97)

0.865 (0.809–0.922)

0.810 (0.706–0.909)

0.904 (0.838–0.963)

0.855 (0.759–0.943)

0.872 (0.800–0.939)

Test group

0.92 (0.84–0.99)

0.869 (0.787–0.951)

0.800 (0.650–0.933)

0.917 (0.811–1.000)

0.870 (0.714–1.000)

0.868 (0.75–0.954)

  1. AUC, area under the ROC curve; CI, confidence interval; PPV, positive predictive value; NPV, negative predictive value; mpMRI, CE-MRI + T2WI + ADC sequences; comprehensive model, mpMRI (ROItumor + ROI3mm) + clinical model.