Skip to main content

Table 2 The performance of three models in training and test cohorts

From: Imaging phenotyping using 18F-FDG PET/CT radiomics to predict micropapillary and solid pattern in lung adenocarcinoma

 

AUC (95%CI)

SEN (%)

SPE (%)

ACC (%)

PPV (%)

NPV (%)

Training cohort

 PET model

0.829 (0.794–0.866)

81.65

71.17

77.00

78.01

75.60

 CT model

0.827 (0.791–0.865)

58.56

94.24

78.40

74.01

89.04

 Combined model

0.870 (0.840–0.902)

79.86

80.18

80.00

83.46

76.07

Test cohort 1

 PET model

0.799 (0.732–0.868)

84.76

69.01

78.41

80.18

75.38

 CT model

0.833 (0.772–0.893)

87.62

67.60

79.55

78.69

80.00

 Combined model

0.859 (0.800–0.918)

89.52

73.24

82.95

83.19

82.54

Test cohort 2

 PET model

0.854 (0.791–0.917)

90.20

70.59

82.35

82.14

82.76

 CT model

0.829 (0.763–0.895)

83.33

72.06

78.82

81.73

74.24

 Combined model

0.880 (0.826–0.934)

86.27

79.41

83.53

86.27

79.41

  1. AUC area under the curve, 95% CI 95% confidence interval, SEN sensitivity, SPE specificity, ACC accuracy, PPV positive predictive value, NPV negative predictive value