Skip to main content

Table 2 Net reclassification improvement analyses between EB and BN2 based on ANN groups

From: Enhanced diagnostic accuracy for quantitative bone scan using an artificial neural network system: a Japanese multi-center database project

 

EB

BN2

0 to 0.24

0.25 to 0.49

0.50 to 0.74

0.75 to 1.00

Total

Metastasis (n = 169): net gain in reclassification proportion = 3.6%, p = 0.38

0 to 0.24

1

3

3

3

10

0.6%

1.8%

1.8%

1.8%

 

0.25 to 0.49

2

2

4

5

13

1.2%

1.2%

2.4%

3.0%

 

0.50 to 0.74

2

1

5

8

16

1.2%

0.6%

3.0%

4.7%

 

0.75 to 1.00

1

3

11

115

130

0.6%

1.8%

6.5%

68.0%

 

Total

6

9

23

131

169

No metastasis (n = 334): net gain in reclassification proportion = −26.0%, p < 0.0001

0 to 0.24

96

31

14

4

145

28.7%

9.3%

4.2%

1.2%

 

0.25 to 0.49

41

14

5

2

62

12.3%

4.2%

1.5%

0.6%

 

0.50 to 0.74

34

19

9

5

67

10.2%

5.7%

2.7%

1.5%

 

0.75 to 1.00

35

11

8

6

60

10.5%

3.3%

2.4%

1.8%

 

Total

206

75

36

17

334

  1. Total NRI = 29.6%, p < 0.0001.