Table 2 Classification performance of the radiomics models in the validation cohort.

From: Prediction of breast cancer molecular subtypes using radiomics signatures of synthetic mammography from digital breast tomosynthesis

  

TN vs non-TN

HER2 vs non-HER2

Luminal vs non-luminal

CC model

AUC

0.819

0.520

0.659

Accuracy

0.817

0.761

0.563

Sensitivity

0.750

0.222

0.440

Specificity

0.831

0.839

0.867

MLO model

AUC

0.791

0.645

0.627

Accuracy

0.718

0.747

0.521

Sensitivity

0.917

0.111

0.480

Specificity

0.678

0.839

0.619

CC + MLO model

AUC

0.838

0.556

0.645

Accuracy

0.803

0.704

0.507

Sensitivity

0.833

0.111

0.440

Specificity

0.797

0.790

0.667

  1. TN triple-negative, AUC area under the receiver operating characteristic curve.
Back to article page