Skip to main content

Table 1 The characteristics of 18,441 participants

From: The prediction of asymptomatic carotid atherosclerosis with electronic health records: a comparative study of six machine learning models

Feature Training set
(n = 12,909)
Testing set
(n = 5532)
Total population
(n = 18,441)
Characteristics    
Age 50.88 (19, 96) 50.81 (18, 93) 50.86 (18, 96)
Age subgroup, y    
 18–64 11,170 (86.6) 4815 (87.0) 15,991 (86.7)
 > 64 1733 (13.4) 717 (13.0) 2450 (13.3)
Gender (male) 7738 (59.9) 3297 (59.6) 11,035 (59.8)
SBP, mmHg 128.69 ± 17.76 128.04 ± 17.43 128.49 ± 17.66
Heart rate, beats/min 75.71 ± 7.80 75.48 ± 7.93 75.64 ± 7.84
Pulse, beats/min 77.40 ± 10.88 76.92 ± 10.94 77.25 ± 10.90
Waistline, cm 83.66 ± 10.10 83.70 ± 9.89 83.67 ± 10.04
Co-morbid conditions    
Hypertension 1502 (11.6) 868 (15.7) 2370 (12.9)
Diabetes mellitus 524 (4.1) 326 (5.9) 850 (4.6)
Hyperlipidemia 177 (1.4) 210 (3.8) 387 (2.1)
Family history 287 (2.2) 235 (4.2) 522 (2.8)
Ever-smoker 563 (4.4) 284 (5.1) 847 (4.6)
Laboratory values    
Glucose, mmol/L 5.42 ± 1.36 5.31 ± 1.36 5.39 ± 1.36
HDL, mmol/L 1.23 ± 0.21 1.21 ± 0.25 1.22 ± 0.22
TC, mmol/L 4.40 ± 0.62 4.40 ± 0.80 4.40 ± 0.68
Total protein, g/L 68.89 ± 4.21 69.15 ± 3.68 68.96 ± 4.06
Albumin, g/L 44.30 ± 2.78 44.21 ± 2.40 44.27 ± 2.67
Albumin/Globulin 1.84 ± 0.28 1.82 ± 0.29 1.84 ± 0.28
γ-GLT, U/L 27.33 ± 26.86 27.12 ± 25.18 27.27 ± 26.37
Platelets, 10^9/L 217.36 ± 58.42 218.57 ± 57.25 217.72 ± 58.07
Carotid atherosclerosis    
Yes 4587 (35.5) 1966 (35.5) 6553 (35.5)
 Age 18–64 y 3246 (25.2) 1418 (25.6) 4670 (25.3)
 Age > 64 y 1335 (10.3) 548 (9.9) 1883 (10.2)
No 8322 (64.5) 3566 (64.5) 11,888 (64.5)
 Age 18–64 y 7924 (61.4) 3397 (61.4) 11,321 (61.4)
 Age > 64 y 398 (3.1) 169 (3.1) 567 (3.1)
  1. Categorical features represented as frequency (%). Continuous features represented as median ± SD, except age, which was median (minimum, maximum). (SBP, systolic blood pressure; HDL, high density lipoprotein; TC, total cholesterol; γ-GLT, γ-glutamyl transpeptidase)