Table 2 Performance comparisons between ultrasound small-window entropy and Nakagami imaging in classifying benign and malignant breast tumors.

From: Small-window parametric imaging based on information entropy for ultrasound tissue characterization

Methodology

Small-window ultrasound entropy imaging

Nakagami statistical parametric imaging

Median (IQR) of the Nakagami parameter

Benign

4.86 (4.57–4.96)

0.59 (0.46–0.67)

Malignant

4.29 (3.87–4.51)

0.38 (0.25–0.53)

Dynamic range of the parameter

1.73–5.09

0.24–0.87

Cutoff value

4.52

0.47

Sensitivity%

76.66%

70.00%

Specificity%

81.81%

69.69%

Accuracy%

79.36%

69.84%

LR+

4.21

2.31

LR−

0.28

0.43

PPV%

79.31

67.74

NPV%

79.41

71.87

AUROC (95% CI)

0.89 (0.80–0.97)

0.79 (0.67–0.90)

  1. LR+: positive likelihood ratio, LR−: negative likelihood ratio, PPV: positive predictive value, NPV: negative predictive value, AUROC: area under the receiver operating characteristics curve.
Back to article page