Skip to main content

Table 2 Specific data for 4 subjects in the training set

From: Opening the black box: interpretable machine learning for predictor finding of metabolic syndrome

Variables

Physical examiner number

3271

6506

25,392

10,557

Gender (0 = female, 1 = male)

0

1

1

0

Age (years)

26

61

45

59

eosinophil percentage

8.3

3.3

2.6

1.6

erythrocyte distribution width coefficient of variation

12.4

14.4

12.9

12.2

creatinine (μmoI/L)

79

52

54

72

uric acid (μmoI/L)

386

260

304

373

glutamyl transpeptidase (U/L)

32

44

16

48

alkaline phosphatase (U/L)

48

115

56

69

previous fatty liver (0 = no, 1 = yes)

0

1

1

1

previous hypertension (0 = no, 1 = yes)

0

0

0

1

previous diabetes (0 = no, 1 = yes)

0

0

0

0

WC (cm)

72

91

84

90

SBP (mmHg)

139

154

122

169

DBP (mmHg)

71

85

83

109

FPG (mmol/L)

4.4

5.13

5.18

6.38

TC (mmol/L)

3.76

5.16

5.19

6.22

TG (mmol/L)

0.77

2.49

1.79

2.98

HDL-C (mmol/L)

1.54

1.22

1.67

1.33

MetS (0 = no, 1 = yes)

0

1

0

1