Skip to main content

Table 3 Diagnostic performances of the radiomic predictive model and the BI-RADS analysis carried out by the radiologist

From: Radiomics model to classify mammary masses using breast DCE-MRI compared to the BI-RADS classification performance

 

Radiomics model performance

BI-RADS performance

Sensibility

100% [85–100]

100% [95–100]

Specificity

33% [10–70]

17% [3–56]

Positive predictive value

84% [65–94]

80% [70–87]

Negative predictive value

100% [34–100]

100% [57–100]

Accuracy

0.85 [0.66–0.96]

0.82 [0.63–0.92]

AUC

0.94 [0.85–1]

0.84 [0.73–0.95]