Skip to main content

Table 4 Prediction performance of C_model, R_model, and RC_model in the training and test sets

From: Clinical parameters combined with radiomics features of PET/CT can predict recurrence in patients with high-risk pediatric neuroblastoma

Set

Model

Sensitivity (95%CI)

Specificity (95%CI)

Accuracy (95%CI)

AUC (95%CI)

Training

C_model

0.645 (0.454–0.808)

0.700 (0.457–0.881)

0.667 (0.521–0.792)

0.744 (0.595–0.874)

 

R_model

0.774 (0.589–0.904)

0.700 (0.457–0.881)

0.745 (0.604–0.857)

0.813 (0.685–0.916)

 

RC_model

0.806 (0.625–0.925)

0.800 (0.563–0.943)

0.804 (0.669–0.902)

0.889 (0.794–0.963)

Test

C_model

0.700 (0.457–0.881)

0.692 (0.386–0.909)

0.697 (0.513–0.844)

0.750 (0.577–0.904)

 

R_model

0.800 (0.563–0.943)

0.769 (0.462–0.950)

0.788 (0.611–0.910)

0.869 (0.715–0.985)

 

RC_model

0.900 (0.683–0.988)

0.769 (0.462–0.950)

0.848 (0.681–0.949)

0.892 (0.758–0.992)

AUC Area under the curve, CI Confidence interval