From: Machine learning methodologies versus cardiovascular risk scores, in predicting disease risk
Variable used in ML | Male | Female |
---|---|---|
Age in years, mean ± SD | 46 ± 13 | 45 ± 14 |
Smoking status at baseline, %(yes) | 44% | 37% |
Years of school mean ± SD | 12.3 ± 3.6 | 12.0 ± 3.8 |
MedDietScore (range 0–55), mean ± SD | 24 ± 5 | 27 ± 7 |
Basic metabolic rate as a proxy of energy expenditure | 1783 ± 228 | 1384 ± 128 |
Body mass index in kg/m2, mean ± SD | 27.3 ± 3.9 | 25.2 ± 4.7 |
Diastolic blood pressure levels in mmHg, mean ± SD | 82 ± 11 | 76 ± 11 |
Systolic blood pressure levels in mmHg, mean ± SD | 127 ± 17 | 118 ± 18 |
History of hypertension (including medication), % | 39% | 24% |
Glucose levels (in mg/dl), mean ± SD | 95 ± 25 | 90 ± 22 |
History of diabetes mellitus (including medication), % | 8% | 6% |
Total cholesterol levels (in mg/dl), mean ± SD | 197 ± 42 | 191 ± 41 |
Triglycerides (in mg/dl), mean ± SD | 140 ± 102 | 98 ± 56 |
History of hypercholesterolemia (including medication), % | 46% | 38% |
Interleukin-6 levels (ng/ml), mean ± SD | 1.5 ± 0.5 | 1.4 ± 0.5 |