Skip to main content

Table 3 Performance comparison of the predictive models in the test cohort

From: RESOLVE-DWI-based deep learning nomogram for prediction of normal-sized lymph node metastasis in cervical cancer: a preliminary study

Model

AUC (95% CI)

P-value

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

ADC

0.719 (0.561–0.845)

0.347

54.6

87.5

60.0

84.8

DWI

0.724 (0.567–0.849)

0.137

63.6

81.3

53.8

86.7

Combined

0.767 (0.613–0.882)

0.427

63.6

81.3

53.8

86.7

Nomogram

0.844 (0.701–0.936)

Reference

72.7

90.6

72.7

90.6

  1. AUC Area under the curve; CI Confidence interval; PPV Positive predictive value; NPV Negative predictive value; ADC Apparent diffusion coefficient; DWI Diffusion-weighted imaging
  2. P < 0.05 indicates statistically significant
Back to article page

BMC Medical Imaging

ISSN: 1471-2342

Contact us