Innovative Integration of Machine Learning Techniques for Early Prediction of Metabolic Syndrome Risk Factors

Over the past two decades, chronic degenerative diseases have risen to prominence in global and national morbidity and mortality statistics. Notably, type 2 diabetes mellitus, arterial hypertension, and metabolic syndrome have been highlighted for their prevalence and have been identified by the World Health Organization (WHO) as potential causes of 50% of worldwide fatalities. Despite increased awareness driven by internet dissemination about risks associated with sedentary lifestyles and poor diets, and the subsequent shift in public perception towards healthier living, it remains a reality that individual concern typically arises following the initial symptomatology of these conditions. In response to this situation, the current study proposes the development of an early warning system, underpinned by advanced machine learning algorithms such as LightGBM, XGBoost, and ensemble methods based on Random Forests that employ gradient boosting techniques to enhance predictive accuracy. This model processes data efficiently, requiring minimal computational resources, to provide personalized risk predictions based on categorical characteristics, as well as biometric and clinical variables. In light of the findings, the deployment of a web platform is anticipated, which will enable individuals to assess their health risk using readily available indicators, thereby promoting greater proactivity in personal health management