Skip to main content

Table 4 Comparison of the proposed method with the traditional methods

From: Segmentation of lung parenchyma in CT images using CNN trained with the clustering algorithm generated dataset

Methods

DSC (vs observer A) (%)

DSC (vs observer B)

Average DSC

Self-adaptability

Iteration

95.40

95.22

95.31

83.33

Improved Ostu

95.35

95.18

95.27

83.68

Watershed

94.72

94.49

94.61

62.50

Region growing

96.65

96.50

96.58

91.32

Proposed method

96.80

96.62

96.71

100