Table 3 Summary of predictive models based on pre-contrast images and percentage change in radiomics features between pre- and post-contrast images

From: Diagnostic value of radiomics and machine learning with dynamic contrast-enhanced magnetic resonance imaging for patients with atypical ductal hyperplasia in predicting malignant upgrade

  AUROC Sensitivity (%) Specificity (%) PPV (%) NPV (%) Accuracy (%)
Pre-contrast 0.514
(0.299–0.728)
15.3
(1.8–42.8)
79.2
(49.2–95.3)
34.8
(11.6–77.3)
48.7
(39.3–56.5)
53.6
(27.5–66.1)
Percentage change 0.540
(0.329–0.752)
22.8
(4.7–50.8)
86.0
(57.2–98.2)
57.5
(22.7–88.4)
53.0
(43.5–60.7)
60.7
(33.9–72.5)
  1. Models were created using gaussian support vector machines (SVMs) and are presented with confidence intervals
  2. AUROC area under the receiver operating curve, NPV negative predictive value, PPV positive predictive value