Skip to main content

Table 4 Diagnostic performance of the Radiology, DECT and DECT–Radiology models in the training and validation cohorts

From: Development and validation of a model for predicting the expression of Ki-67 in pancreatic ductal adenocarcinoma with radiological features and dual-energy computed tomography quantitative parameters

Models

Training cohort

Validation cohort

AUC (95% CI)

SEN

SPE

DeLong

AUC (95% CI)

SEN

SPE

DeLong

Radiology model

0.705 (0.590–0.820)

0.826

0.587

0.001a

0.669 (0.499–0.839)

0.786

0.552

0.093a

DECT model

0.884 (0.816–0.952)

0.913

0.779

0.300b

0.835 (0.715–0.955)

0.786

0.828

0.514b

DECT–Radiology model

0.905 (0.841–0.968)

0.826

0.883

 < 0.001c

0.865 (0.757–0.972)

0.857

0.793

0.003c

  1. AUC area under the curve, CI confidence interval, DECT dual-energy computed tomography, SEN sensitivity, SPE specificity
  2. aRadiology model versus DECT model
  3. bDECT model versus DECT–Radiology model
  4. cDECT–Radiology versus Radiology model