| Issue/Objective |
Renal transplantation is a critical treatment for end-stage renal disease, but graft failure remains a significant concern. Accurate prediction of graft survival is crucial to identify high-risk patients. This study aimed to develop prognostic models for predicting renal graft survival and compare the performance of statistical and machine learning models. |
| Methodology/Approach |
This study analyzed data from 278 kidney transplant recipients at the Ethiopian National Kidney Transplantation Center between September 2015 and February 2022. To overcome data limitations, we employed a comprehensive methodological approach. This included SMOTE resampling to address class imbalance, rigorous cross-validation for robust model evaluation, and pre-feature selection to identify the most relevant predictors. We then compared the performance of several advanced survival models: standard and penalized Cox regression, Random Survival Forest, and Stochastic Gradient Boosting. Hyperparameter tuning was used to optimize model performance. Prognostic predictors were ultimately selected based on both statistical significance and variable importance scores across the models. |
| Results |
The median graft survival time was 33 months with a 3-year horizon graft survival rate of 0.911. The Stochastic Gradient Boosting (SGB) model demonstrated the best discrimination and calibration performance, with a C-index of 0.943 and a Brier score of 0.000351. The Ridge-based Cox model closely followed the SGB model's prediction performance with better interpretability. The key prognostic predictors of graft survival included an episode of acute and chronic rejections, post-transplant urological complications, post-transplant nonadherence, blood urea nitrogen level, post-transplant regular exercise, and marital status. |
| Discussion/Conclusion |
The Stochastic Gradient Boosting model demonstrated the highest predictive performance, while the Ridge-Cox model offered better interpretability with a comparable performance. Clinicians should consider the trade-off between prediction accuracy and interpretability when selecting a model. Incorporating these findings into clinical practice can improve risk stratification and personalized management strategies for kidney transplant recipients. |
| Presenters and Affiliations |
Getahun Mulugeta Awoke Debre Berhan University |
