Skip to main content

Table 2 Efficacy of different models in predicting pathological responses after NAC in advanced gastric cancer in the training and validation groups

From: Intratumoral and peritumoral radiomics predict pathological response after neoadjuvant chemotherapy against advanced gastric cancer

Model

Intratumoral model

Peritumoral model

Combined model

Clinical model

Intratumoral-clinical model

Peritumoral-clinical model

Combined-clinical model

Training group

 AUC

0.943

0.846

0.949

0.756

0.955

0.898

0.960

 95% CI

0.896–0.974

0.781–0.898

0.903–0.978

0.682–0.820

0.910–0.981

0.840–0.940

0.917–0.984

 p value

 < 0.0001

 < 0.0001

 < 0.0001

 < 0.0001

 < 0.0001

 < 0.0001

 < 0.0001

 ACC (%)

89.44

74.53

90.06

73.91

93.79

83.85

91.93

 SEN(%)

86.73

63.27

93.88

81.63

93.88

86.73

93.88

 SPE(%)

93.65

92.06

84.13

61.90

93.65

79.37

88.89

 PPV(%)

95.51

92.54

90.20

76.92

95.83

86.73

92.93

 NPV(%)

81.94

61.70

89.83

68.42

90.77

79.37

90.32

Validation group

 AUC

0.778

0.701

0.815

0.672

0.808

0.759

0.843

 95% CI

0.663–0.869

0.579–0.804

0.705–0.898

0.549–0.779

0.696–0.892

0.641–0.853

0.736–0.919

 p value

 < 0.0001

0.002

 < 0.0001

0.0093

 < 0.0001

 < 0.0001

 < 0.0001

 ACC (%)

75.71

70.00

82.86

62.86

74.29

77.14

81.43

 SEN(%)

78.57

80.95

95.24

57.14

69.05

83.33

80.95

 SPE(%)

71.43

53.57

64.29

71.43

82.14

67.86

82.14

 PPV(%)

80.49

72.34

80.00

75.00

85.29

79.55

87.18

 NPV(%)

68.97

65.22

90.00

52.63

63.89

73.08

74.19

  1. AUC Area under the curve, CI Confidence interval, ACC Accuracy, SEN Sensitivity, SPE Specificity, PPV Positive predictive value, NPV Negative predictive value