Table 4 Comparison of descriptive statistics from paU-Net and nnU-Net.

From: Fully automated deep learning based auto-contouring of liver segments and spleen on contrast-enhanced CT images

 

Seg 1 N = 392

Seg 2 N = 40

Seg 3 N = 40

Seg 4 N = 40

Seg 5–8 N = 40

paU-Net1

nnU- Net1

paU-Net1

nnU- Net1

paU-Net1

nnU- Net1

paU-Net1

nnU- Net1

paU-Net1

nnU- Net1

DSC

0.90

0.93

0.82

0.86

0.89

0.91

0.86

0.91

0.97

0.97

0.88

0.90

0.81

0.83

0.87

0.89

0.83

0.88

0.96

0.97

0.07

0.07

0.08

0.09

0.09

0.10

0.11

0.08

0.03

0.03

0.96

0.98

0.92

0.94

0.94

0.96

0.95

0.97

0.99

0.99

0.71

0.70

0.56

0.51

0.44

0.36

0.46

0.56

0.88

0.86

 

**

*

**

**

**

\({{\text{HD}}}_{95}^{3}\)

4

3.1

8

8.0

8

6.6

9

6.5

6

4.5

5

4.6

11

9.5

9

8.0

25

8.3

10

7.6

5

6.1

8

6.0

5

5.5

34

6.7

11

8.8

31

37.1

34

33.4

26

28.9

118

36.7

46

47.8

1

0.8

4

3.1

4

2.5

4

2.1

1

0.7

 

*

ns

ns

**

ns

\({{\text{HD}}}_{{\text{A}}}^{3}\)

0.20

0.14

0.60

0.48

0.38

0.31

0.49

0.24

0.09

0.06

0.35

0.30

0.79

0.69

0.57

0.49

1.18

0.47

0.22

0.18

0.50

0.61

0.61

0.62

0.81

0.87

1.78

0.55

0.37

0.29

2.99

3.85

2.55

2.92

4.56

5.37

9.43

2.71

1.85

1.54

0.06

0.02

0.19

0.11

0.14

0.08

0.11

0.04

0.01

0.01

 

**

ns

*

**

ns

PDV

10

7

16

11

6

5

7

7

3

2

14

8

22

21

9

8

12

8

3

4

15

10

22

35

8

8

13

7

3

4

80

45

123

173

39

27

54

30

19

22

0

0

1

0

0

0

0

0

0

0

 

*

ns

ns

**

ns

  1. 1Data in each cell is organized as row 1 = Median, row 2 = Mean, row 3 = Standard deviation, row 4 = Max, row 5 = Min row row 6 = significance level, P \(\le 0.05\) = *, P \(\le 0.01\) = **, P \(>\) 0.05 = ns.
  2. 2One of the patients was removed because paU-Net failed to predict segment 1.
  3. 3\({{\text{HD}}}_{{\text{A}}}\) and \({{\text{HD}}}_{95}\) average and 95% Hausdorff distance (mm); PDV = percent difference in volume.