Análisis predictivo del riesgo de hipertensión en adultos mexicanos basado en indicadores nutricionales y calóricos

This article develops a predictive analysis of hypertension risk in Mexican adults based on nutritional and caloric indicators. Hypertension, a condition with serious health implications, requires the identification of predictive risk factors for its prevention and effective management. Several machine learning models were evaluated, with the Random Forest model standing out for its high accuracy and robustness, while XGBoost excelled in efficiency with large datasets. In contrast, the Naive Bayes model showed the lowest performance. Additionally, the study emphasizes the importance of macronutrients and total caloric intake in predicting hypertension, with proteins, carbohydrates, and lipids being relevant risk factors, especially in young adults in Mexico. This finding highlights the need to integrate multiple nutritional factors in risk assessment.