Skip to main content

Table 3 Discrimination performance of predict models for predicting pCR status in breast cancer patients

From: Development and validation of a radiopathomic model for predicting pathologic complete response to neoadjuvant chemotherapy in breast cancer patients

Training Set

AUC (95%CI)

SEN (%)

SPE (%)

ACC (%)

PPV (%)

NPV (%)

HER2

0.686(0.611–0.762)

63.93

73.04

69.67

60.93

75.82

RS

0.858(0.801–0.916)

88.52

67.02

75.48

63.52

90.00

PS

0.803(0.734–0.872)

65.57

81.91

75.48

70.17

78.57

DLPS

0.862(0.805–0.920)

85.24

74.46

78.70

68.42

88.60

DLRPM

0.933(0.895–0.971)

90.16

85.10

87.09

79.71

93.02

Validation Set

AUC (95%CI)

SEN (%)

SPE (%)

ACC (%)

PPV (%)

NPV (%)

HER2

0.738(0.617–0.858)

76.19

71.42

73.21

61.53

83.33

RS

0.821(0.700–0.942)

85.71

77.14

80.35

69.23

90.00

PS

0.766(0.629–0.903)

52.38

91.42

76.78

78.57

76.19

DLPS

0.804(0.683–0.925)

61.90

91.42

80.35

81.25

80.00

DLRPM

0.927(0.858–0.996)

94.28

76.19

87.50

71.25

92.25

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