Skip to main content

Table 2 Classification performance for the baseline methods

From: Automatic small bowel tumor diagnosis by using multi-scale wavelet-based analysis in wireless capsule endoscopy images

 

[29]

[16]

Classification Vector

2 nd order moments

2 nd and 3 rd order moments

Histogram based features

N TU based features

Specificity (μ ± σ%)

88.0 ± 0.5

90.1 ± 0.3

84.7 ± 0.7

88.2 ± 0.3

Sensitivity (μ ± σ%)

88.1 ± 0.4

91.2 ± 0.3

85.5 ± 0.7

89.1 ± 0.5

  1. This table presents the classification performance of the methods used as baseline reference to assess the performance of the proposed algorithm.