Skip to main content

Table 3 Prediction performance of CIAKI models in the internal and external validation cohorts

From: A novel explainable online calculator for contrast-induced AKI in diabetics: a multi-centre validation and prospective evaluation study

Classifier

AUC (%)

(95% CI)

Accuracy (%)

(95% CI)

Sensitivity (%)

(95% CI)

Specificity (%)

(95% CI)

PPV (%)

(95% CI)

NPV (%)

(95% CI)

F1 score (%)

(95% CI)

Internal validation performance

 Mehran score

65.41

(60.85–69.93)

67.57

(64.53–70.78)

50.47

(42.57–58.68)

71.34

(67.92–74.59)

27.98

(22.63–33.17)

86.72

(83.95–89.43)

36.00

(30.07–41.82)

 LR

79.90

(75.95–83.32)

78.21

(75.34–81.08)

57.01

(49.04–64.55)

82.89

(80.00–85.66)

42.36

(35.62–48.82)

89.73

(87.39–92.01)

48.61

(41.92–54.62)

 LASSO + LR

79.92

(75.91–83.44)

78.21

(75.68–80.74)

56.07

(47.97–63.46)

83.09

(80.21–85.71)

42.25

(35.17–48.63)

89.56

(87.17–91.96)

48.19

(41.30–54.10)

 SVM

80.73

(76.78–83.99)

79.05

(76.35–81.76)

55.14

(47.32–62.83)

84.33

(81.57–86.95)

43.70

(36.57–50.72)

89.50

(87.26–91.81)

48.76

(41.88–55.23)

 RF

79.76

(76.07–83.06)

78.89

(76.18–81.59)

53.27

(45.61–61.11)

84.54

(81.78–86.95)

43.18

(36.17–50.00)

89.13

(86.75–91.54)

47.70

(40.98–53.85)

 GBDT

80.28

(76.49–83.98)

80.41

(77.70–83.11)

57.01

(49.51–64.81)

85.57

(82.85–88.02)

46.56

(39.31–53.78)

90.02

(87.78–92.36)

51.26

(44.64–57.52)

 XGBT

81.59

(77.71–85.33)

81.08

(78.55–83.78)

54.21

(46.23–62.26)

87.01

(84.66–89.41)

47.93

(40.59–55.36)

89.60

(87.32–91.97)

50.88

(44.05–57.26)

External validation performance

 Mehran score

65.59

(60.18–71.10)

73.83

(70.76–76.71)

37.50

(28.57–46.84)

79.96

(76.89–83.05)

24.00

(17.74–30.66)

88.34

(85.75–90.76)

29.27

(22.55–36.36)

 LR

76.94

(71.84–82.03)

72.02

(68.95–75.09)

71.25

(62.82–79.22)

72.15

(68.92–75.48)

30.16

(25.00–35.67)

93.70

(91.42–95.71)

42.38

(36.36–48.63)

 LASSO + LR

79.31

(74.40–84.17)

72.20

(69.13–75.27)

73.75

(65.17–82.05)

71.94

(68.78–75.26)

30.73

(25.77–36.04)

94.20

(91.98–96.21)

43.38

(37.36–49.64)

 SVM

76.47

(71.14–81.58)

70.58

(67.51–73.65)

65.00

(56.32–73.56)

71.52

(68.35–74.79)

27.81

(22.70–33.16)

92.37

(90.00–94.47)

38.95

(32.96–44.85)

 RF

80.10

(75.33–84.64)

72.56

(69.49–75.63)

75.00

(66.67–82.89)

72.15

(68.90–75.48)

31.25

(26.02–36.95)

94.48

(92.39–96.46)

44.12

(38.02–50.52)

 GBDT

78.16

(73.27–83.01)

72.38

(69.31–75.27)

76.25

(68.24–83.75)

71.73

(68.39–74.95)

31.28

(26.09–36.59)

94.71

(92.76–96.47)

44.36

(38.38–50.49)

 XGBT

81.65

(77.04–86.09)

75.99

(73.10–79.06)

72.50

(64.00–80.82)

76.58

(73.24–79.74)

34.32

(28.57–40.34)

94.29

(92.19–96.16)

46.59

(40.17–52.80)

  1. LR logistic regression, LASSO least absolute shrinkage and selection operator, SVM support vector machine, RF random forest, GBDT gradient boosted decision trees, XGBT extreme gradient boosting trees, AUC area under the curve, PPV positive predictive value, NPV negative predictive value, 95% CI 95% confidence interval