Skip to main content

Table 1 Evaluation of the tumour extraction algorithm in BCC histology slides

From: Performance of a simple chromatin-rich segmentation algorithm in quantifying basal cell carcinoma from histology images

Tumours

n

Sensitivity (%)

Specificity (%)

PPV (%)

NPV (%)

Infiltrative

10

78.95

87.25

71.15

87.88

Superficial

10

98.13

82.50

34.34

99.74

Nodular

10

95.82

89.31

84.53

94.96

All

30

90.97

86.35

63.34

94.19

  1. PPV: Positive Predictive Value = [True Positive Pixels/(True Positive Pixels + False Positive Pixels)] · 100%; NPV: Negative Predictive Value = [True Negative Pixels/(True Negative Pixels + False Negative Pixels)] · 100%; Sensitivity = [True Positive Pixels/(True Positive Pixels + False Negative Pixels)] · 100%; Specificity = [True Negative Pixels/(True negative Pixels + False Positive Pixels)] · 100%
Back to article page