Machine Learning Approach to Extract Diagnostic and Prognostic Thresholds: Application in Prognosis of Cardiovascular Mortality

Table 1

Baseline characteristics.

	Frequency in percent or median
Demographic variables
Men, % (𝑛)	32.1 (177)
Age, years	67.1±8
Race, % (𝑛)
Mixed	73.1 (404)
Caucasian	22.2 (122)
African-Venezuelan	4 (22)
Natives	0.5 (3)
Use of antihypertensive drugs, % (𝑛)	30.9 (170)
Use of anti-diabetic drugs, % (𝑛)	11.1 (61)
History of cardiovascular disease, % (𝑛)	11.5 (63)
Diagnosis of diabetes mellitus, % (𝑛)	18.1 (100)
Lifestyle, physical and lipid factors
Smoking current status, % (𝑛)	15.6 (86)
Drinking current status, % (𝑛)	31.6 (174)
Body max index, kg/m^2	27.1±5.6
Total serum cholesterol, mmol/L	5.5±1.3
24-hour ambulatory measurements
Systolic blood pressure, mm Hg	133.8±16.6
Diastolic blood pressure, mm Hg	76.1±10
Heart rate, bpm	73.7±9.8