%0 Journal Article
%T Predicting Increased Blood Pressure Using Machine Learning
%A Hudson Fernandes Golino
%A Liliany Souza de Brito Amaral
%A Stenio Fernando Pimentel Duarte
%A Cristiano Mauro Assis Gomes
%A Telma de Jesus Soares
%A Luciana Araujo dos Reis
%A Joselito Santos
%J Journal of Obesity
%D 2014
%I Hindawi Publishing Corporation
%R 10.1155/2014/637635
%X The present study investigates the prediction of increased blood pressure by body mass index (BMI), waist (WC) and hip circumference (HC), and waist hip ratio (WHR) using a machine learning technique named classification tree. Data were collected from 400 college students (56.3% women) from 16 to 63 years old. Fifteen trees were calculated in the training group for each sex, using different numbers and combinations of predictors. The result shows that for women BMI, WC, and WHR are the combination that produces the best prediction, since it has the lowest deviance (87.42), misclassification (.19), and the higher pseudo (.43). This model presented a sensitivity of 80.86% and specificity of 81.22% in the training set and, respectively, 45.65% and 65.15% in the test sample. For men BMI, WC, HC, and WHC showed the best prediction with the lowest deviance (57.25), misclassification (.16), and the higher pseudo (.46). This model had a sensitivity of 72% and specificity of 86.25% in the training set and, respectively, 58.38% and 69.70% in the test set. Finally, the result from the classification tree analysis was compared with traditional logistic regression, indicating that the former outperformed the latter in terms of predictive power. 1. Introduction Obesity (body mass index > 29.9£¿kg/m2) has been considered a global public health problem due to its high prevalence and high morbidity [1]. In fact, the prevalence of obesity has increased substantially, both in developed and in under development countries. In the United States, for example, it is estimated that 35.5% of women and 32.2% of adult men present obesity [2]. The Brazilian Institute of Geography and Statistics (IBGE) indicates that 50.1% of men and 48% of women have overweight (25£¿kg/m2 ¡Ü BMI < 29.9£¿kg/m2), while 12.4% of men and 16.9% of women are suffering from obesity in Brazil [3]. The high risk attributed to obesity is related particularly to its association with increased risk factors for cardiovascular disease, notably hypertension [4, 5]. In order to adopt early preventive/therapeutic actions to minimize the risk of cardiovascular events in obese individuals, methods that can predict hypertension using low cost procedures are necessary, especially in underdeveloped and in developing countries. Body mass index (BMI), waist circumference (WC), hip circumference (HC), and waist-hip ratio (WHR) are among the most practical and cost effective measures for evaluation of obesity, with the advantage that both WC and WHR present positive correlations with the amount of visceral fat, and together
%U http://www.hindawi.com/journals/jobe/2014/637635/