Predictive models of anthropometric parameters for primary screening of sarcopenia based on Machine Learning
Revista Mexicana de Ingeniería Biomédica · 29.06.2026 · 47 (2) · e1565
DOI: 10.17488/rmib.47.2.1565 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista Mexicana de Ingeniería Biomédica, online erschienen am 29.06.2026. Verfasst von Santiago Arceo Díaz und 6 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
This work reports a free-access primary screening system for detecting sarcopenia risk in older Mexican adults, using machine learning and anthropometric variables obtained through accessible instruments such as measuring tapes. An observational, retrospective, and analytical study was conducted based on records from beneficiaries of the Mexican Social Security Institute from the year 2019, with a sample of 1,678 participants. The models, developed using data from individuals without comorbidities, followed a structured machine learning workflow that included data preprocessing, variable transformation and clustering, and supervised classification using decision-tree-based models. The optimal variable combinations for men and women achieved F1-scores above 0.94, accurately classifying the risk levels of sarcopenia and severe sarcopenia. The current models need to be expanded to include individuals with comorbidities such as type 2 diabetes, hypertension, and arthritis, which have been associated with greater muscle mass loss. This proposal does not replace clinical diagnostic testing but serves as a complementary tool to rule out low-risk individuals and prioritize specialized evaluation for those who may be affected by sarcopenia.
Zitieren
APA
Díaz, S. A., Barrios, E. E. B., Trujillo-Trujillo, X. A. R., Sánchez-García, S., Barrios, J. A. B., Silva, M. R. S. R., & Viera, M. H. (2026). Predictive models of anthropometric parameters for primary screening of sarcopenia based on Machine Learning. Revista Mexicana de Ingeniería Biomédica, 47(2), e1565. https://doi.org/10.17488/rmib.47.2.1565
Vancouver
Díaz SA, Barrios EEB, Trujillo-Trujillo XAR, Sánchez-García S, Barrios JAB, Silva MRSR, et al. Predictive models of anthropometric parameters for primary screening of sarcopenia based on Machine Learning. Revista Mexicana de Ingeniería Biomédica. 2026;47(2):e1565. doi:10.17488/rmib.47.2.1565
BibTeX
@article{Daz2026,
author = {Díaz, Santiago Arceo and Barrios, Elena Elsa Bricio and Trujillo-Trujillo, Xóchitl Angélica Rosio and Sánchez-García, Sergio and Barrios, Jaime Alberto Bricio and Silva, Mónica Rios Silva Ríos and Viera, Miguel Huerta},
title = {Predictive models of anthropometric parameters for primary screening of sarcopenia based on Machine Learning},
journal = {Revista Mexicana de Ingeniería Biomédica},
year = {2026},
volume = {47},
number = {2},
pages = {e1565},
doi = {10.17488/rmib.47.2.1565}
}
EUVIMED