Salgado C, Gonzalez Cohens F, Vera FA, Ruiz R, Velasquez JD, Gonzalez FM. Prediction of graft outcomes after kidney transplantation: When standard statistics compare to machine learning techniques. World J Nephrol 2026; 15(1): 116879 [DOI: 10.5527/wjn.v15.i1.116879]
Corresponding Author of This Article
Fernando M Gonzalez, MD, Department of Nephrology, Faculty of Medicine, Universidad de Chile, Avenida Salvador 486, Providencia, Santiago 7500922, Chile. fgonzalf@uc.cl
Research Domain of This Article
Transplantation
Article-Type of This Article
Retrospective Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
World J Nephrol. Mar 25, 2026; 15(1): 116879 Published online Mar 25, 2026. doi: 10.5527/wjn.v15.i1.116879
Prediction of graft outcomes after kidney transplantation: When standard statistics compare to machine learning techniques
Carolina Salgado, Francisca Gonzalez Cohens, Felipe A Vera, Rocío Ruiz, Juan D Velasquez, Fernando M Gonzalez
Carolina Salgado, Francisca Gonzalez Cohens, Felipe A Vera, Rocío Ruiz, Juan D Velasquez, Web Intelligence Centre, Faculty of Physics and Mathematical Sciences, Universidad de Chile, Santiago 7500922, Chile
Fernando M Gonzalez, Department of Nephrology, Faculty of Medicine, Universidad de Chile, Santiago 7500922, Chile
Author contributions: Salgado C built and ran all the models; Gonzalez Cohens F and Gonzalez FM revised the methodology, model performance, and interpretability; Salgado C, Gonzalez Cohens F, and Gonzalez FM wrote the manuscript; Vera FA, Ruiz R, and Velasquez JD reviewed, edited, and approved the manuscript.
Supported by a public grant from Agencia Nacional De Investigación Y Desarrollo, No. ID23I10232.
Institutional review board statement: The Comité Ético Científico del Servicio de Salud Metropolitano Oriente reviewed and approved the study (No. 29092020).
Informed consent statement: Written informed consent was obtained from all kidney transplant recipients upon their inclusion on the transplant centers’ waiting lists, authorizing the use of their anonymized clinical data.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
Data sharing statement: Data can be shared upon request.
Corresponding author: Fernando M Gonzalez, MD, Department of Nephrology, Faculty of Medicine, Universidad de Chile, Avenida Salvador 486, Providencia, Santiago 7500922, Chile. fgonzalf@uc.cl
Received: November 24, 2025 Revised: January 12, 2026 Accepted: February 9, 2026 Published online: March 25, 2026 Processing time: 111 Days and 7.3 Hours
Core Tip
Core Tip: Machine learning (ML) is increasingly used in kidney transplantation research, including predicting delayed graft function. This study compares six ML models with logit across four donor, transplant, and recipient variable combinations. The dataset comprises 44.7% delayed graft function-positive cases. All methods have similar performances, with accuracies between 58%-70%. Important predictors included donor creatinine, age, and mean blood pressure, cold-ischemia time, and recipient smoking condition. Although ML approaches slightly outperformed logit, overall performance remained modest, likely due to limited sample-size. Further research should define dataset scale and quality for ML to become a primary analytic tool for predicting kidney transplant outcomes.
