BODY COMPOSITION PREDICTS HYPERTENSION USING MACHINE LEARNING METHODS: A COHORT STUDY

Abstract

Objective: Hypertension is one of the most important and preventable causes of cardiovascular disease (CVD), stroke, chronic kidney disease, and dementia which caused approximately 8.5 million deaths in 2015, in low & middle income countries. Hypertension depends on well known risk factors such as age, gender, family history, smoking, alcohol consumption, central obesity, overweight and physical inactivity. Evidence shows that body fat distribution is a more vital determinant of cardiovascular morbidity and mortality than increased fat mass. In the past few years, growing number of researchers have used machine learning and data mining algorithms to diagnose and treat health conditions such as heart11 and brain12 diseases. Their non-invasive nature and accuracy have enabled health professionals to quickly identify at-risk individuals and use more efficient preventive and managing strategies. In this study, we used machine learning approaches to investigate whether BIA-derived body composition indices predict hypertension in a cohort of patients. Design and method: Data from a cohort study was used. Body composition analysis was done using bioelectrical impedance analysis (BIA); weight, basal metabolic rate, total and regional fat percentage (FATP), and total and regional fat-free mass (FFM) were measured. We used machine learning methods such as Support Vector Classifer, Decision Tree, Stochastic Gradient Descend Classifer, Logistic Regression, Gaussian Naïve Bayes, K-Nearest Neighbor, Multi-Layer Perceptron, Random Forest, Gradient Boosting, Histogram-based Gradient Boosting, Bagging, Extra Tree, Ada Boost, Voting, and Stacking to classify the investigated cases and fnd the most relevant features to hypertension. Results: 4663 records were included (2156 were male, 1099 with hypertension, with the age range of 35–70 years old). Arm FFM, basal metabolic rate, total FFM, Trunk FFM, leg FFM, and male gender were inversely associated with hypertension, but total FATP, arm FATP, leg FATP, older age, trunk FATP, and female gender were directly associated with hypertension. AutoMLP, stacking and voting methods had the best performance for hypertension prediction achieving an accuracy rate of 90%, 84% and 83%, respectively. Conclusions: By using machine learning methods, we found that BIA-derived body composition indices predict hypertension with acceptable accuracy.