Table 5 Comparison of volume overlap error (VOE) for atlas-based segmentation against deep-learning-based segmentation with four organs (heart, liver, kidney, stomach). Averages and standard deviations are listed for ten test cases

# 1

8.49

6.83

10.22

10.58

12.88

10.31

30.15

18.37

84.83

31.01

# 2

11.68

13.24

9.95

7.81

17.62

18.38

14.39

19.62

45.96

39.11

# 3

7.34

11.17

8.39

13.44

12.42

11.07

63.90

12.42

65.73

34.86

# 4

11.17

12.73

6.73

7.34

8.86

8.54

8.34

10.31

81.17

54.12

# 5

13.87

11.89

10.29

11.64

9.39

7.59

20.76

35.07

72.42

31.16

# 6

26.60

11.95

34.32

15.32

15.17

15.66

14.32

17.84

55.58

35.68

# 7

24.39

6.66

20.41

11.02

8.10

9.03

9.99

12.00

32.21

41.44

# 8

19.34

12.12

13.29

10.30

15.27

15.41

16.58

12.53

39.02

23.24

# 9

7.74

10.34

13.14

12.07

40.66

13.06

20.15

14.59

85.51

42.75

# 10

21.04

11.48

8.37

8.68

34.68

12.87

57.73

10.39

63.92

41.97

Avg SD

15.17

10.84

13.51

10.82

17.51

12.19

25.63

16.31

62.64

37.53

6.77

2.18

7.83

2.36

10.57

3.33

18.57

7.01

18.10

7.99