A machine learning approach to personalized predictors of dyslipidemia - a cohort study

Aug 5, 2026

Upload by Finn

A machine learning study builds personalized predictors for dyslipidemia by analyzing a Mexico City cohort of 2,621 participants aged 20–50 with and without dyslipidemia. Variable Importance Measures from Random Forest, XGBoost, and Gradient Boosting Machine guide feature selection, while SMOTE addresses class imbalance. Model comparison using balanced accuracy, sensitivity, and specificity identifies an optimal attribute subset and highlights top risk-related features, supporting early identification and prevention.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Introduction

Cohort background and motivation

Methods

Dataset and machine learning workflow

Feature selection and class imbalance handling

Results

Model performance and selected attributes

Discussion

Key contributing features and implications

TYPE Original Research PUBLISHED 20 September 2023 DOI 10. 3389/fpubh.2023.1213926 OPEN ACCESS EDITED BY Zhendong Liu, Shandong First Medical University, China REVIEWED BY Suman Kundu, Vanderbilt University Medical Center, United States Zonglin He, Hong Kong University of Science and Technology, Hong Kong SAR, China *CORRESPONDENCE Enrique Hernández-Lemus [ehernandez@inmegen.gob.mx](ehernandez@inmegen.gob.mx)[ ](ehernandez@inmegen.gob.mx)[Guadalupe Guti](Guadalupe Guti)é[rrez-Esparza](rrez-Esparza) [ggutierreze@conacyt.mx](ggutierreze@conacyt.mx) †These authors share ﬁrst authorship RECEIVED 28 April 2023 ACCEPTED 23 August 2023 PUBLISHED 20 September 2023 CITATION Gutiérrez-Esparza G, Pulido T, Martínez-García M, Ramírez-delReal T, Groves-Miralrio LE, Márquez-Murillo MF, Amezcua-Guerra LM, Vargas-Alarcón G and Hernández-Lemus E (2023) A machine learning approach to personalized predictors of dyslipidemia: a cohort study. Front. Public Health 11:1213926 . doi: 10.3389/fpubh.2023.1213926 COPYRIGHT © 2023 Gutiérrez-Esparza, Pulido, Martínez-García, Ramírez-delReal, Groves-Miralrio, Márquez-Murillo, Amezcua-Guerra, Vargas-Alarcón and Hernández-Lemus. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. A machine learning approach to personalized predictors of dyslipidemia: a cohort study Guadalupe Gutiérrez-Esparza1,2*†, Tomas Pulido2†, Mireya Martínez-García3 , Tania Ramírez-delReal1,4 , Lucero E. Groves-Miralrio3 , Manlio F. Márquez-Murillo5 , Luis M. Amezcua-Guerra3 , Gilberto Vargas-Alarcón6 and Enrique Hernández-Lemus7,8* 1 Researcher for Mexico CONAHCYT, National Council of Humanities, Sciences, and Technologies, Mexico City, Mexico, 2 Clinical Research, National Institute of Cardiology “Ignacio Chávez”, Mexico City, Mexico, 3 Department of Immunology, National Institute of Cardiology “Ignacio Chávez”, Mexico City, Mexico, 4 Center for Research in Geospatial Information Sciences, Aguascalientes, Mexico, 5 Department of Electrocardiology, National Institute of Cardiology “Ignacio Chávez”, Mexico City, Mexico, 6 Department of Molecular Biology and Endocrinology, National Institute of Cardiology “Ignacio Chávez”, Mexico City, Mexico, 7 Computational Genomics Division, National Institute of Genomic Medicine, Mexico City, Mexico, 8 Center for Complexity Sciences, Universidad Nacional Autónoma de México, Mexico City, Mexico Introduction: Mexico ranks second in the global prevalence of obesity in the adult population, which increases the probability of developing dyslipidemia. Dyslipidemia is closely related to cardiovascular diseases, which are the leading cause of death in the country. Therefore, developing tools that facilitate the prediction of dyslipidemias is essential for prevention and early treatment. Methods: In this study, we utilized a dataset from a Mexico City cohort consisting of 2,621 participants, men and women aged between 20 and 50 years, with and without some type of dyslipidemia. Our primary objective was to identify potential factors associated with di􀀀erent types of dyslipidemia in both men and women. Machine learning algorithms were employed to achieve this goal. To facilitate feature selection, we applied the Variable Importance Measures (VIM) of Random Forest (RF), XGBoost, and Gradient Boosting Machine (GBM) . Additionally, to address class imbalance, we employed Synthetic Minority Over-sampling Technique (SMOTE) for dataset resampling. The dataset encompassed anthropometric measurements, biochemical tests, dietary intake, family health history, and ot

FAQs

What data and population were used in the cohort study?

The study used a Mexico City cohort dataset with 2,621 participants (men and women) aged 20–50, including individuals with and without dyslipidemia.

Which machine learning methods were used for feature selection and prediction?

Variable Importance Measures from Random Forest, XGBoost, and Gradient Boosting Machine were applied for feature selection, combined with classifier performance evaluation.

How did the study handle class imbalance?

Synthetic Minority Over-sampling Technique (SMOTE) was used to resample the dataset and address class imbalance.

Aug 5, 2026

Upload by Finn