Health Care Technologies and Analytics: Transforming Modern Healthcare

Abstract

Obesity is a growing global health problem that increases vulnerability to many chronic diseases. Several chronic diseases are attributed to obesity including diabetes and cardiovascular disorders. Machine learning methods have potential in predicting the risk of obesity using lifestyle and physiological parameters. There is a need for efficient and accurate methods to predict the risk of obesity. This paper compares logistic regression, decision tree, random forest, support vector machine (SVM), and gradient boosting methods using a voting classifier to predict the risk of obesity. A healthcare-related dataset was used in this study and the performance was evaluated using accuracy and classification metrics. The results show that using ensemble learning models improve the prediction performance especially in the case of using multiple models. This paper contributes to justifying the use of ensemble methods in the field of healthcare analytics. Gradient Boosting (LightGBM) had the best performance, and thus it is the most suitable model for this task. Gradient Boosting (LightGBM) had the best performance, and thus it is the most suitable model for this task.  The technique can be adopted to develop efficient and reliable decision-making applications in healthcare.