Diagnosing Vascular Aging Based on Macro and Micronutrients Using Ensemble Machine Learning

Carmen Patino-Alonso (), Marta Gómez-Sánchez, Leticia Gómez-Sánchez, Emiliano Rodríguez-Sánchez, Cristina Agudo-Conde, Luis García-Ortiz and Manuel A Gómez-Marcos
Additional contact information
Carmen Patino-Alonso: Department of Statistics, University of Salamanca, Campus Miguel de Unamuno, C/Alfonso X el Sabio s/n, 37007 Salamanca, Spain
Marta Gómez-Sánchez: Primary Care Research Unit of Salamanca (APISAL), Biomedical Research Institute of Salamanca (IBSAL), Avenida de Portugal 83, 37005 Salamanca, Spain
Leticia Gómez-Sánchez: Primary Care Research Unit of Salamanca (APISAL), Biomedical Research Institute of Salamanca (IBSAL), Avenida de Portugal 83, 37005 Salamanca, Spain
Emiliano Rodríguez-Sánchez: Primary Care Research Unit of Salamanca (APISAL), Biomedical Research Institute of Salamanca (IBSAL), Avenida de Portugal 83, 37005 Salamanca, Spain
Cristina Agudo-Conde: Primary Care Research Unit of Salamanca (APISAL), Biomedical Research Institute of Salamanca (IBSAL), Avenida de Portugal 83, 37005 Salamanca, Spain
Luis García-Ortiz: Primary Care Research Unit of Salamanca (APISAL), Biomedical Research Institute of Salamanca (IBSAL), Avenida de Portugal 83, 37005 Salamanca, Spain
Manuel A Gómez-Marcos: Primary Care Research Unit of Salamanca (APISAL), Biomedical Research Institute of Salamanca (IBSAL), Avenida de Portugal 83, 37005 Salamanca, Spain

Mathematics, 2023, vol. 11, issue 7, 1-18

Abstract: The influence of dietary components on vascular dysfunction and aging is unclear. This study therefore aims to propose a model to predict the influence of macro and micronutrients on accelerated vascular aging in a Spanish population without previous cardiovascular disease. This cross-sectional study involved a total of 501 individuals aged between 35 and 75 years. Carotid-femoral pulse wave velocity (cfPWV) was measured using a Sphygmo Cor ® device. Carotid intima-media thickness (IMTc) was measured using a Sonosite Micromax ® ultrasound machine. The Vascular Aging Index (VAI) was estimated according to VAI = (LN (1.09) × 10 cIMT + LN (1.14) × cfPWV) 39.1 + 4.76. Vascular aging was defined considering the presence of a vascular lesion and the p75 by age and sex of VAI following two steps: Step 1: subjects were labelled as early vascular aging (EVA) if they had a peripheral arterial disease or carotid artery lesion. Step 2: they were classified as EVA if the VAI value was >p75 and as normal vascular aging (NVA) if it was ≤p75. To predict the model, we used machine learning algorithms to analyse the association between macro and micronutrients and vascular aging. In this article, we proposed the AdXGRA model, a stacked ensemble learning model for diagnosing vascular aging from macro and micronutrients. The proposed model uses four classifiers, AdaBoost (ADB), extreme gradient boosting (XGB), generalized linear model (GLM), and random forest (RF) at the first level, and then combines their predictions by using a second-level multilayer perceptron (MLP) classifier to achieve better performance. The model obtained an accuracy of 68.75% in prediction, with a sensitivity of 66.67% and a specificity of 68.79%. The seven main variables related to EVA in the proposed model were sodium, waist circumference, polyunsaturated fatty acids (PUFA), monounsaturated fatty acids (MUFA), total protein, calcium, and potassium. These results suggest that total protein, PUFA, and MUFA are the macronutrients, and calcium and potassium are the micronutrients related to EVA.

Keywords: machine learning technique; stacking classifiers; macronutrient; micronutrient; accelerated vascular aging (search for similar items in EconPapers)
JEL-codes: C (search for similar items in EconPapers)
Date: 2023
References: View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.mdpi.com/2227-7390/11/7/1645/pdf (application/pdf)
https://www.mdpi.com/2227-7390/11/7/1645/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jmathe:v:11:y:2023:i:7:p:1645-:d:1110329

Access Statistics for this article

Mathematics is currently edited by Ms. Emma He

More articles in Mathematics from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().