Machine Learning for Predicting Delayed Graft Function and Graft Survival After Kidney Transplantation: A Systematic Review and Meta-Analysis
- 1MedStar Georgetown Transplant Institute, Department of Surgery, MedStar Georgetown University Hospital, Georgetown University School of Medicine, 3800 Reservoir Road NW, Washington, DC 20007, USA
Abstract
Background/Objectives: Machine-learning (ML) models are promoted for predicting delayed graft function (DGF) and graft survival after kidney transplantation; we examined whether they outperform regression and how well either has been validated. Methods: We searched PubMed/MEDLINE, Embase, and Web of Science through 28 June 2026, searched citations for named DGF models, extracted data in duplicate, assessed risk of bias using the Prediction Model Risk of Bias ASsessment Tool (PROBAST), and pooled only external validations of the same published model. Results: We included 217 reports (145 on DGF and 91 on graft survival, 19 on both); across seven external cohorts, the Irish 2010 DGF model had a pooled area under the curve (AUC) of 0.73 (95% confidence interval 0.69–0.77), the median difference in AUC or C-statistic between ML and regression in 28 same-cohort comparisons was 0.01 (interquartile range −0.02 to 0.03), and 208 of the 212 non-pediatric reports were at high risk of bias. Conclusions: ML has not been shown to predict DGF or graft survival better than regression, and the comparisons are too heterogeneous to show that it is worse; a model should inform organ acceptance or recipient counseling only after it has shown discrimination and calibration in an independent cohort.
Related
- Predicting MCI-to-Alzheimer’s Conversion: A Systematic Benchmark of Classical, Machine Learning, and Deep Survival Models
- Development and temporal validation of nomograms and machine learning models for predicting overall and cancer-specific survival in primary distal cholangiocarcinoma: a SEER-based study
- Bedside Bioimpedance and Simple Functional Tests Identify Muscle Function Impairment Predicting Overall Survival in Chemotherapy-treated Metastatic Urothelial Carcinoma: A Retrospective Pilot Study
- Feasibility of a machine learning model for predicting weight gain after kidney transplantation machine learning and weight gain after kidney transplantation
- Survival and reactivation risk after heart transplantation in Chagas heart disease: a systematic review and individual patient data meta-analysis
BibTeX
@article{gogna2026machine,
title = {{Machine Learning for Predicting Delayed Graft Function and Graft Survival After Kidney Transplantation: A Systematic Review and Meta-Analysis}},
author = {Shekhar Gogna and Piyush Gupta and David Walls and Reginald Gohh},
journal = {Journal of Clinical Medicine},
year = {2026},
month = nov,
doi = {10.3390/jcm15218535},
url = {https://www.mdpi.com/2077-0383/15/21/8535},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Machine Learning for Predicting Delayed Graft Function and Graft Survival After Kidney Transplantation: A Systematic Review and Meta-Analysis
AU - Shekhar Gogna
AU - Piyush Gupta
AU - David Walls
AU - Reginald Gohh
PY - 2026
DA - 2026/11/02
JO - Journal of Clinical Medicine
DO - 10.3390/jcm15218535
UR - https://www.mdpi.com/2077-0383/15/21/8535
AB - Background/Objectives: Machine-learning (ML) models are promoted for predicting delayed graft function (DGF) and graft survival after kidney transplantation; we examined whether they outperform regression and how well either has been validated. Methods: We searched PubMed/MEDLINE, Embase, and Web of Science through 28 June 2026, searched citations for named DGF models, extracted data in duplicate, assessed risk of bias using the Prediction Model Risk of Bias ASsessment Tool (PROBAST), and pooled only external validations of the same published model. Results: We included 217 reports (145 on DGF and 91 on graft survival, 19 on both); across seven external cohorts, the Irish 2010 DGF model had a pooled area under the curve (AUC) of 0.73 (95% confidence interval 0.69–0.77), the median difference in AUC or C-statistic between ML and regression in 28 same-cohort comparisons was 0.01 (interquartile range −0.02 to 0.03), and 208 of the 212 non-pediatric reports were at high risk of bias. Conclusions: ML has not been shown to predict DGF or graft survival better than regression, and the comparisons are too heterogeneous to show that it is worse; a model should inform organ acceptance or recipient counseling only after it has shown discrimination and calibration in an independent cohort.
SN - 2077-0383
ER -