Skip to main content

Table 1 Clinical characteristics of the patients in training and testing cohorts

From: The added value of radiomics from dual-energy spectral CT derived iodine-based material decomposition images in predicting histological grade of gastric cancer

Characteristics

Training set (n = 72)

Testing set (n = 31)

Low-grade

(n = 28)

High-grade

(n = 44)

P value

Low-grade

(n = 12)

High-grade

(n = 19)

P value

Age

70.29 ± 8.19

66.43 ± 11.12

0.119

65.33 ± 11.48

64.11 ± 11.72

0.777

Gender

  

0.714

  

0.178

 Female

9

16

 

1

7

 

 Male

19

28

 

11

12

 

Tumor long axis

4.53 ± 2.73

5.39 ± 2.85

0.209

3.60 ± 1.71

5.66 ± 3.55

0.039

pT stage

  

0.005

  

0.025

 T1

5

0

 

5

0

 

 T2

4

5

 

0

1

 

 T3

5

4

 

2

4

 

 T4

14

35

 

5

14

 

pN stage

  

 < 0.001

  

 < 0.001

 N0

11

6

 

6

1

 

 N1

5

2

 

2

1

 

 N2

7

17

 

3

4

 

 N3

5

29

 

1

13

 

Location

  

0.115

  

0.278

 Upper 1/3

10

10

 

3

4

 

 Middle 1/3

10

14

 

5

5

 

 Lower 1/3

7

16

 

4

7

 

 Multiple

1

4

 

0

3

 

AFP

  

1.000

  

1.000

 Normal

26

41

 

12

19

 

 Abnormal

2

3

 

0

0

 

CEA

  

0.147

  

1.000

 Normal

23

42

 

12

19

 

 Abnormal

5

2

 

0

0

 

CA125

  

0.884

  

0.510

 Normal

25

41

 

12

17

 

 Abnormal

3

3

 

0

2

 

CA19-9

  

0.553

  

0.409

 Normal

22

37

 

7

15

 

 Abnormal

6

7

 

5

4

 

CA72-4

  

0.620

  

0.510

 Normal

23

34

 

12

17

 

 Abnormal

5

10

 

0

2

 

CA153

  

1.000

  

1.000

 Normal

27

43

 

12

19

 

 Abnormal

1

1

 

0

0

 
Back to article page

BMC Medical Imaging

ISSN: 1471-2342

Contact us