Skip to main content

Table 2 Clinical characteristics of the training set validation set

From: A machine learning-based severity stratification tool for high altitude pulmonary edema

Characteristic

test, N = 1001

train, N = 2361

p-value2

Sex

76 (76%)

190 (81%)

0.4

Age

37 (26, 44)

31 (26, 42)

0.095

Height

169 (165, 172)

170 (165, 173)

0.056

Weight

72 (62, 78)

74 (65, 77)

0.4

BMI

24.6 (22.7, 26.2)

25.0 (23.1, 26.4)

0.4

Fatigue

20 (20%)

62 (26%)

0.2

Loss of taste

24 (24%)

53 (22%)

0.8

Nausea

71 (71%)

194 (82%)

0.03

Sleep disorder

20 (20%)

52 (22%)

0.7

Dizzy

70 (70%)

143 (61%)

0.1

Cough

83 (83%)

204 (86%)

0.4

Short breath

52 (52%)

123 (52%)

> 0.9

Dyspnea

25 (25%)

53 (22%)

0.6

Palpitations

21 (21%)

41 (17%)

0.4

Bluishlips

19 (19%)

30 (13%)

0.14

Sputum

  

0.7

 no

94 (94%)

214 (91%)

 

 yes

6 (6.0%)

21 (8.9%)

 

Lungrales

  

0.4

 None

16 (16%)

52 (22%)

 

 Unilateral or bilateral lung floor rales (small amount)

55 (55%)

113 (48%)

 

 Unilateral or bilateral middle and lower lobes of the lungs (moderate amount)

24 (24%)

53 (22%)

 

 Large amount of rales (full lungs)

5 (5.0%)

18 (7.6%)

 

SPO2

75 (65, 84)

76 (68, 85)

0.4

SBP

120 (115, 132)

120 (111, 130)

0.3

DBP

80 (70, 86)

80 (70, 88)

> 0.9

ct

  

0.5

 Normal

1 (1.0%)

8 (3.4%)

 

 Double lung/texture

7 (7.0%)

13 (5.5%)

 

 Bipulmonary lower field inner bands/patchy/bipulmonary lower field inner bands/flocculent

26 (26%)

52 (22%)

 

 Others

66 (66%)

163 (69%)

 

Group

  

> 0.9

 Mild

20 (20%)

47 (20%)

 

 Moderate

72 (72%)

168 (71%)

 

 Sever

8 (8.0%)

21 (8.9%)

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