Skip to main content

Table 4 Performance evaluation of ADC parameters and nomogram model for differentiating occult IF and bone metastasis

From: Radiation-induced occult insufficiency fracture or bone metastasis after radiotherapy for cervical cancer? The nomogram based on quantitative apparent diffusion coefficients for discrimination

Parameters

Cut off value

AUC (95% CI)

Sensitivity

Specificity

PPV

NPV

ADC mean (×10−6 mm2/s)

990

0.853 (0.756,0.950)

85.4% (41/48)

85.2% (23/27)

91.1% (41/45)

76.7% (23/30)

ADC min (×10−6 mm2/s)

828

0.743 (0.625, 0.861)

70.8% (34/48)

77.8% (21/27)

85.0% (34/40)

60.0% (21/35)

ADC max (×10−6 mm2/s)

1044

0.837 (0.732, 0.942)

89.6% (43/48)

77.8% (21/27)

87.8% (43/49)

80.8% (21/26)

ADC std (×10−6 mm2/s)

99.5

0.778 (0.670, 0.886)

66.7% (32/48)

88.9% (24/27)

91.4% (32/35)

60.0% (24/40)

ADC ratio (Lesion/normal bone)

3.75

0.853 (0.756, 0.950)

85.4% (41/48)

85.1% (23/27)

91.1% (41/45)

76.7% (23/30)

ADC ratio (Lesion/lateral muscle)

0.77

0.887 (0.793, 0.980)

95.8% (46/48)

81.5% (22/27)

90.2% (46/51)

91.7% (22/24)

Nomogram model (score)

0.97

0.921 (0.848, 0.995)

91.7% (44/48)

96.3% (26/27)

97.8% (44/45)

86.7%(26/30)

  1. AUC area under the ROC curve, PPV positive predictive value, NPV negative predictive value