Predictive Modeling of Blood Pressure Categories: Integrating Demographic and Dietary Factors for Personalized Management

This article is a preprint and has not been peer-reviewed.

For citation:
Show BibTeX format

Yuan, L. "Predictive Modeling of Blood Pressure Categories: Integrating Demographic and Dietary Factors for Personalized Management." GitData Archive, vol. 2024, no. 01, Jan. 2024, https://archive.gd.edu.kg/20231224040449/

PREPRINTS.RU: https://preprints.ru/article/1216/

Abstract:

This study delves into predictive modeling of blood pressure levels, focusing on the United States, addressing the global health concern of hypertension. Mainly utilizing demographic and dietary data from the Centers for Disease Control and Prevention (CDC) National Health and Nutrition Examination Survey (NHANES) 2017-2018, aims to craft personalized management strategies. Drawing on research emphasizing the multifaceted determinants of hypertension, we leverage the multinomial regression model with lasso regularization as a baseline. Furthermore, the study advances to the eXtreme Gradient Boosting (XGBoost) algorithm, achieving a better performance than multinomial regression. Evaluation metrics include accuracy and Area Under the ROC Curve (ROC-AUC) in a 10-fold cross validation framework. The study provides possible personal blood pressure management solution.

License:

This work is licensed under CC BY-NC-SA 4.0.