Development of Health Parameter Model for Risk Prediction of CVD Using SVM

Table 4

Regression coefficients and associated statistics obtained from BMES dataset for male and female subjects.

Feature	Male				Female
	Coefficient		Odds ratio	95% confidence interval	Coefficient		Odds ratio	95% confidence interval
Age (per 1 year)	0.0144	<0.00001	1.015	(1.012, 1.017)	0.0110	<0.00001	1.011	(1.009, 1.012)
Body mass index (per 1 kg/m^2)	0.0084	0.0024	1.008	(1.002, 1.013)	0.0018	0.2317	1.002	(0.998, 1.004)
Current smoker (versus past or never smoker)	0.0911	0.0005	1.095	(1.042, 1.153)	0.0749	0.0003	1.078	(1.034, 1.122)
Systolic blood pressure (per 1 mmHg)	0.0003	0.5700	1.000	(0.999, 1.001)	0.0003	0.3416	1.000	(0.999, 1.0009)
Medication for hypertension (versus no medication for hypertension)	−0.0042	0.8589	0.995	(0.953, 1.039)	0.0218	0.1521	1.022	(0.992, 1.0219)
Diabetes (versus no diabetes)	0.0460	0.1834	1.047	(0.979, 1.119)	0.0080	0.7852	1.008	(0.951, 1.067)
Total cholesterol (per 1 mmol/L)	0.0131	0.1615	1.013	(0.995, 1.032)	0.0016	0.8052	1.002	(0.989, 1.014)
High-density lipoprotein cholesterol (per 1 mmol/L)	0.0202	0.4524	1.020	(0.969, 1.074)	0.0046	0.7828	1.005	(0.972, 1.037)

Logistic regression constant for male = −5.70203; logistic regression constant for female = −5.30218.