Skip to main content

Table 4 Performance of the established RSs based on primary lesion and SLN, and their combination for HER2-positive and HER2-negative subtypes

From: DCE-MRI radiomics of primary breast lesions combined with ipsilateral axillary lymph nodes for predicting efficacy of NAT

  

Training cohort

Validation cohort

Test cohort

Subtypes

Model

AUC (95% CI)

ACC

SEN

SPE

AUC (95% CI)

ACC

SEN

SPE

AUC (95% CI)

ACC

SEN

SPE

HER2+

RS-Primary

0.721 (0.702–0.781)

0.710

0.735

0.691

0.714 (0.681–0.734)

0.653

0.683

0.801

0.709 (0.653–0.777)

0.653

0.645

0.760

RS-SLN

0.700 (0.610–0.761)

0.661

0.633

0.740

0.691 (0.582–0.804)

0.591

0.739

0.610

0.683 (0.571–0.810)

0.602

0.619

0.896

RS-Com

0.812 (0.720–0.891)

0.833

0.891

0.705

0.801 (0.710–0.903)

0.783

0.729

0.842

0.780 (0.631–0.871)

0.730

0.619

0.845

HER2-

RS-Primary

0.719 (0.640–0.877)

0.690

0.702

0.679

0.710 (0.654–0.812)

0.652

0.680

0.792

0.704 (0.581–0.822)

0.634

0.620

0.770

RS-SLN

0.703 (0.621–0.802)

0.682

0.610

0.733

0.687 (0.565–0.780)

0.581

0.742

0.589

0.669 (0.591–0.820)

0.621

0.602

0.798

RS-Com

0.782 (0.692–0.901)

0.811

0.831

0.693

0.781 (0.621–0.904)

0.762

0.710

0.862

0.776 (0.672–0.892)

0.734

0.608

0.831

  1. AUC, area under the receiver operating characteristic curve; CI, confidence interval; ACC, accuracy; SEN, sensitivity; SPE, specificity; SLN, suspicious lymph node