Table 3 Radiomics and radiomics-clinical model performance evaluation.

From: Radiomics and outcome prediction to antiangiogenic treatment in advanced gastroenteropancreatic neuroendocrine tumours: findings from the phase II TALENT trial

 

AUC [95% CI]

p value

Accuracy (%)

Sensitivity (%)

Specificity (%)

Radiomics model

 Training

0.75 [0.60–0.90]

<0.001

78

76

79

 Validation

0.67 [0.41–0.92]

0.114

55

55

56

 Test

0.67 [0.43–0.91]

0.09

63

67

58

Radiomics clinical model

 Training

0.79 [0.64–0.93]

<0.001

78

90

67

 Validation

0.72 [0.48–0.95]

0.12

70

45

100

 Test

0.58 [0.33–0.82]

0.276

54

17

92

  1. Table containing the area under the curve (AUC), accuracy, sensitivity and specificity for the Youden’s cut-off 0.49 for the radiomics-only model and 0.36 for the radiomics-clinical model.
Back to article page