Skip to main content

Table 6 Subgroup analysis of accuracy between CMGsNET and other four nets in segmentation of severe diseases

From: Utility of a novel integrated deep convolutional neural network for the segmentation of hip joint from computed tomography images in the preoperative planning of total hip arthroplasty

Disease

Dependent variable

(I) NET

(J) NET

Mean difference (I − J)

Std. error

Sig

95% confidence interval

Lower bound

Upper bound

Multiple comparisons

Least significance difference (LSD)

ONFH III

ASD

CMG net

FCN

− 0.19

0.04

< 0.05

− 0.27

− 0.12

2D UNET

− 0.27

0.04

< 0.05

− 0.34

− 0.19

2.5D UNET

− 0.25

0.04

< 0.05

− 0.33

− 0.17

3D UNET

− 0.22

0.04

< 0.05

− 0.30

− 0.14

DOC

CMG net

FCN

0.10

0.03

< 0.05

0.04

0.15

2D UNET

0.04

0.03

0.12

− 0.01

0.10

2.5D UNET

0.20

0.03

< 0.05

0.15

0.25

3D UNET

0.21

0.03

< 0.05

0.16

0.26

HD

CMG net

FCN

− 1.10

0.46

< 0.05

− 2.01

− 0.19

2D UNET

− 0.33

0.46

0.47

− 1.24

0.58

2.5D UNET

− 5.67

0.46

< 0.05

− 6.58

− 4.76

3D UNET

− 6.67

0.46

< 0.05

− 7.58

− 5.76

ONFH IV

ASD

CMG net

FCN

0.00

0.07

0.99

− 0.14

0.14

2D UNET

− 0.09

0.07

0.20

− 0.23

0.05

2.5D UNET

− 0.09

0.07

0.19

− 0.23

0.04

3D UNET

− 0.10

0.07

0.15

− 0.24

0.04

DOC

CMG net

FCN

0.14

0.03

< 0.05

0.09

0.19

2D UNET

0.05

0.03

0.05

0.00

0.10

2.5D UNET

0.21

0.03

< 0.05

0.16

0.26

3D UNET

0.23

0.03

< 0.05

0.18

0.28

HD

CMG net

FCN

− 1.31

0.40

< 0.05

− 2.11

− 0.52

2D UNET

− 0.77

0.40

0.06

− 1.57

0.02

2.5D UNET

− 5.10

0.40

< 0.05

− 5.89

− 4.30

3D UNET

− 7.13

0.40

< 0.05

− 7.92

− 6.34

DDH III

ASD

CMG net

FCN

− 0.12

0.05

< 0.05

− 0.23

− 0.02

2D UNET

− 0.18

0.05

< 0.05

− 0.28

− 0.07

2.5D UNET

− 0.37

0.05

< 0.05

− 0.47

− 0.26

3D UNET

− 0.13

0.05

< 0.05

− 0.24

− 0.03

DOC

CMG net

FCN

0.18

0.03

< 0.05

0.11

0.24

2D UNET

0.07

0.03

< 0.05

0.01

0.13

2.5D UNET

0.22

0.03

< 0.05

0.15

0.28

3D UNET

0.26

0.03

< 0.05

0.20

0.33

HD

CMG net

FCN

− 1.46

0.67

< 0.05

− 2.80

− 0.11

2D UNET

− 0.15

0.67

0.83

− 1.49

1.20

2.5D UNET

− 5.06

0.67

< 0.05

− 6.41

− 3.72

3D UNET

− 6.60

0.67

< 0.05

− 7.94

− 5.25

DDH IV

ASD

CMG net

FCN

− 0.21

0.07

< 0.05

− 0.34

− 0.08

2D UNET

− 0.20

0.07

< 0.05

− 0.34

− 0.07

2.5D UNET

− 0.25

0.07

< 0.05

− 0.38

− 0.12

3D UNET

− 0.17

0.07

< 0.05

− 0.30

− 0.04

DOC

CMG net

FCN

0.13

0.04

< 0.05

0.05

0.20

2D UNET

0.12

0.04

< 0.05

0.05

0.19

2.5D UNET

0.28

0.04

< 0.05

0.21

0.35

3D UNET

0.28

0.04

< 0.05

0.20

0.35

HD

CMG net

FCN

− 1.29

0.67

0.06

− 2.64

0.07

2D UNET

− 1.50

0.67

< 0.05

− 2.85

− 0.14

2.5D UNET

− 5.47

0.67

< 0.05

− 6.83

− 4.12

3D UNET

− 6.39

0.67

< 0.05

− 7.74

− 5.03

FHF III

ASD

CMG net

FCN

− 0.21

0.02

< 0.05

− 0.26

− 0.16

2D UNET

− 0.24

0.02

< 0.05

− 0.29

− 0.20

2.5D UNET

− 0.24

0.02

< 0.05

− 0.28

− 0.19

3D UNET

− 0.24

0.02

< 0.05

− 0.29

− 0.19

DOC

CMG net

FCN

0.11

0.02

< 0.05

0.08

0.14

2D UNET

0.04

0.02

< 0.05

0.01

0.08

2.5D UNET

0.20

0.02

< 0.05

0.17

0.23

3D UNET

0.18

0.02

< 0.05

0.15

0.21

HD

CMG net

FCN

− 1.21

0.28

< 0.05

− 1.76

− 0.67

2D UNET

0.21

0.28

0.45

− 0.34

0.75

2.5D UNET

− 4.53

0.28

< 0.05

− 5.08

− 3.99

3D UNET

− 6.38

0.28

< 0.05

− 6.93

− 5.84

FHF IV

ASD

CMG net

FCN

− 0.19

0.04

< 0.05

− 0.27

− 0.12

2D UNET

− 0.27

0.04

< 0.05

− 0.34

− 0.19

2.5D UNET

− 0.25

0.04

< 0.05

− 0.33

− 0.17

3D UNET

− 0.22

0.04

< 0.05

− 0.30

− 0.14

DOC

CMG net

FCN

0.10

0.03

< 0.05

0.04

0.15

2D UNET

0.04

0.03

< 0.05

− 0.01

0.10

2.5D UNET

0.20

0.03

< 0.05

0.15

0.25

3D UNET

0.21

0.03

< 0.05

0.16

0.26

HD

CMG net

FCN

− 1.10

0.46

< 0.05

− 2.01

− 0.19

2D UNET

− 0.33

0.46

0.47

− 1.24

0.58

2.5D UNET

− 5.67

0.46

< 0.05

− 6.58

− 4.76

3D UNET

− 6.67

0.46

< 0.05

− 7.58

− 5.76