Skip to main content

Table 3 Results of ROC curve analysis

From: CT imaging-based histogram features for prediction of EGFR mutation status of bone metastases in patients with primary lung adenocarcinoma

Feature

Range

Skewness

Quantile 0.975

Combination

(R + S)

Combination

(R + Q)

Combination

(S + Q)

Combination

(R + S + Q)

AUC

0.765

0.699

0.749

0.783

0.769

0.750

0.783

Threshold

0.501

0.496

0.417

0.460

0.531

0.519

0.462

Specificity

0.708

0.909

0.750

0.788

0.848

0.788

0.788

Sensitivity

0.788

0.417

0.667

0.708

0.625

0.667

0.708

95% CI

Lower bound

0.630

0.561

0.618

0.661

0.641

0.618

0.661

Upper bound

0.899

0.838

0.879

0.905

0.897

0.882

0.905

  1. Footnote: R, range; S, skewness; Q, quantile 0.975