Skip to main content

Table 2 Diagnostic accuracy for significant features differentiating patients with pathologic complete response (pCR) versus non-pCR patients

From: Textural radiomic features and time-intensity curve data analysis by dynamic contrast-enhanced MRI for early prediction of breast cancer therapy response: preliminary data

  

p value*

ROC-AUC

Sensitivity

Specificity

PPV

NPV

Accuracy

Cutoff

Textural features

Δ Entropy

0.024

0.71

0.67

0.73

0.56

0.81

0.71

3.78

Δ LRE

0.021

0.71

0.73

0.70

0.55

0.84

0.71

0.57

Δ Busyness

0.020

0.72

0.67

0.80

0.63

0.83

0.76

34.38

Dynamic features

Δ MSD

0.013

0.74

0.67

0.83

0.67

0.83

0.78

27.74

Δ WIS

< 0.001

0.73

0.60

0.87

0.69

0.81

0.78

73.62

Δ WOS

0.012

0.86

0.87

0.80

0.68

0.92

0.82

24.42

SIS

< 0.001

0.93

0.93

0.87

0.78

0.96

0.89

56.47

  1. LRE Long-run emphasis, MSD Maximum signal difference, NPV Negative predictive value, PPV Positive predictive value, ROC-AUC Receiver operating characteristic area under the curve, SIS Standardised index of shape, WIS Washin, WOS Washout slope
  2. *Kruskal-Wallis test