Feasibility of a machine learning model for predicting weight gain after kidney transplantation machine learning and weight gain after kidney transplantation

Júlia de Melo Cardoso de Freitas, Pedro Lemos Ballester, Elis Forcellini Pedrollo, Camila Corrêa, Rafaella Câmara Rocha Menezes, Cristiane Bauermann Leitão and Gabriela Corrêa Souza

PLOS ONE, 2026, vol. 21, issue 10, 1-12

Abstract: Background: Excessive weight gain after kidney transplantation adversely affects patient and graft outcomes. Machine Learning has been increasingly used to improve risk prediction across a range of transplant-related outcomes, yet its application to post-transplant weight trajectories remains unexplored. This study is the first to evaluate the feasibility of using Machine Learning algorithms to identify predictors of weight gain after kidney transplantation. Methods: This retrospective cohort study analyzed 374 adult kidney transplant recipients. We developed an Elastic Net regression model with 10-fold cross-validation to predict body weight variation at 12 months post-transplant. The analysis screened a comprehensive set of demographic and clinical variables, including recipient characteristics, donor type, dialysis duration, and pre-transplant hypertension, to identify predictors within the training (75%) and testing (25%) datasets. Results: Mean weight gain was 4.0 ± 7.4 kg. The Elastic Net model retained four predictors in the final model: female sex, deceased donor source, age, and pre-transplant body mass index. However, the model demonstrated limited predictive performance (R2 = 0.10; Mean Absolute Error = 8.7%), indicating that baseline variables explain only a small fraction of the observed variability in weight change. Conclusions: This pioneering study establishes an analytical benchmark and reveals the informational limits of standard clinical datasets for predicting post-transplant weight gain. While Machine Learning application is feasible, current demographic and clinical variables are insufficient for precise individual prediction. Future models must integrate more granular behavioral and pharmacological data to achieve clinically actionable risk stratification.

Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0360501 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 60501&type=printable (application/pdf)

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:plo:pone00:0360501

DOI: 10.1371/journal.pone.0360501

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().