BACKGROUND
Kidney transplantation is the preferred treatment option for patients with end-stage renal disease. To maximize patient and graft survival, the allocation of donor organs to potential recipients requires careful consideration.
OBJECTIVE
To develop an innovative technological solution to enable better prediction of kidney transplant survival for each potential donor-recipient pair.
METHODS
We used de-identified data on past organ donors, recipients and transplant outcomes in the United States from the Scientific Registry of Transplant Recipients (SRTR). To predict transplant outcomes for potential donor-recipient pairs, we used several survival analysis models, including regression analysis (Cox Proportional Hazards), Random Survival Forests (RSF) and several artificial neural networks (DeepSurv, DeepHit, Recurrent Neural Networks (RNN)). We evaluated the performance of each model on their ability to predict the probability of graft survival after kidney transplantation from deceased donors. Three metrics were employed: the C-index, the Integrated Brier Score and the Integrated Calibration Index (ICI), along with calibration plots.
RESULTS
Based on the C-index metrics, the neural network-based models (DeepSurv, DeepHit, RNN) had better discriminative ability than the Cox model and RSF (0.650, 0.661, 0.659 vs 0.646 and 0.644, correspondingly). The proposed RNN model offered a compromise between the good discriminative ability and calibration and was implemented in a technological solution of TRL-4.
CONCLUSIONS
Our technological solution based on RNN model can effectively predict kidney transplant survival and provide support for medical professionals and candidate recipients in determining the most optimal donor-recipient pair.
CLINICALTRIAL
Not applicable.
Using machine learning and an ensemble of methods to predict kidney transplant survival
