Skip to main content

Table 1 clinical characteristics of training and test set at baseline

From: Predicting micropapillary or solid pattern of lung adenocarcinoma with CT-based radiomics, conventional radiographic and clinical features

Characteristic

test, N = 1051

train, N = 2461

p-value2

age

63 (56, 67)

62 (55, 67)

0.14

gender

  

> 0.9

 F

53 (50%)

123 (50%)

 

 M

52 (50%)

123 (50%)

 

max diameter(mm)

37 (24, 50)

36 (25, 48)

0.5

smoking history

40 (38%)

89 (36%)

0.7

lung cancer family history

12 (11%)

24 (9.8%)

0.6

ProGRP

44 (37, 56)

43 (38, 56)

> 0.9

CEA

3 (2, 6)

3 (2, 6)

0.7

cyfra

2.20 (1.70, 3.40)

2.20 (1.70, 3.38)

0.9

NSE

13.7 (12.4, 15.3)

13.3 (12.2, 15.1)

0.3

SCCA

0.80 (0.60, 0.90)

0.80 (0.60, 0.98)

> 0.9

type

  

0.6

 CTR < 0.5

11 (10%)

20 (8.1%)

 

 CTR > 0.5

5 (4.8%)

6 (2.4%)

 

 pure-GGO

5 (4.8%)

14 (5.7%)

 

 solid

84 (80%)

206 (84%)

 

lobular

42 (40%)

92 (37%)

0.6

spiculation

52 (50%)

149 (61%)

0.055

pleural traction

27 (26%)

63 (26%)

> 0.9

air bronchus

21 (20%)

54 (22%)

0.7

vessel

43 (41%)

104 (42%)

0.8

hollow

17 (16%)

33 (13%)

0.5

smooth edge

6 (5.7%)

13 (5.3%)

0.9

adjacent to pleural

15 (14%)

30 (12%)

0.6

  1. 1Median (IQR); n (%)
  2. 2Wilcoxon rank sum test; Pearson’s Chi-squared test; Fisher’s exact test