Predicting the outcome of renal transplantation

OBJECTIVE Renal transplantation has dramatically improved the survival rate of hemodialysis patients. However, with a growing proportion of marginal organs and improved immunosuppression, it is necessary to verify that the established allocation system, mostly based on human leukocyte antigen matching, still meets today's needs. The authors turn to machine-learning techniques to predict, from donor-recipient data, the estimated glomerular filtration rate (eGFR) of the recipient 1 year after transplantation. DESIGN The patient's eGFR was predicted using donor-recipient characteristics available at the time of transplantation. Donors' data were obtained from Eurotransplant's database, while recipients' details were retrieved from Charité Campus Virchow-Klinikum's database. A total of 707 renal transplantations from cadaveric donors were included. MEASUREMENTS Two separate datasets were created, taking features with <10% missing values for one and <50% missing values for the other. Four established regressors were run on both datasets, with and without feature selection. RESULTS The authors obtained a Pearson correlation coefficient between predicted and real eGFR (COR) of 0.48. The best model for the dataset was a Gaussian support vector machine with recursive feature elimination on the more inclusive dataset. All results are available at http://transplant.molgen.mpg.de/. LIMITATIONS For now, missing values in the data must be predicted and filled in. The performance is not as high as hoped, but the dataset seems to be the main cause. CONCLUSIONS Predicting the outcome is possible with the dataset at hand (COR=0.48). Valuable features include age and creatinine levels of the donor, as well as sex and weight of the recipient.

[1]  A. Wiseman,et al.  Long-Term Impact of Donor-Recipient Size Mismatching in Deceased Donor Kidney Transplantation and in Expanded Criteria Donor Recipients , 2010, Transplantation.

[2]  R. Ploeg,et al.  The Influence of Deceased Donor Age and Old-for-Old Allocation on Kidney Transplant Outcome , 2009, Transplantation.

[3]  C. Cardella,et al.  Impact of Deceased Donor Diabetes Mellitus on Kidney Transplant Outcomes: A Propensity Score-Matched Study , 2009, Transplantation.

[4]  L. Resende,et al.  Impact of donor age on renal allograft function and survival. , 2009, Transplantation proceedings.

[5]  John F. Hurdle,et al.  Single and multiple time-point prediction models in kidney transplant outcomes , 2008, J. Biomed. Informatics.

[6]  Mohamed Ghoneim,et al.  Prediction of Graft Survival of Living-Donor Kidney Transplantation: Nomograms or Artificial Neural Networks? , 2008, Transplantation.

[7]  N. Ahmad,et al.  Influence of number of retransplants on renal graft outcome. , 2008, Transplantation proceedings.

[8]  Sergey Krikov,et al.  Predicting Kidney Transplant Survival Using Tree-Based Modeling , 2007, ASAIO journal.

[9]  Gunnar Rätsch,et al.  Large Scale Multiple Kernel Learning , 2006, J. Mach. Learn. Res..

[10]  G. Mayer,et al.  Eurotransplant kidney allocation system (ETKAS): rationale and implementation. , 2006, Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association.

[11]  Richard Simon,et al.  Bias in error estimation when using cross-validation for model selection , 2006, BMC Bioinformatics.

[12]  Yong Mao,et al.  Parameters selection in gene selection using Gaussian kernel support vector machines by genetic algorithm. , 2005, Journal of Zhejiang University. Science. B.

[13]  C. Hawley,et al.  Impact of obesity on renal transplant outcomes , 2005, Nephrology.

[14]  Dharmendra Sharma,et al.  Use of Artificial Neural Networks in the Prediction of the Kidney Transplant Outcomes , 2004, KES.

[15]  Nello Cristianini,et al.  Learning the Kernel Matrix with Semidefinite Programming , 2002, J. Mach. Learn. Res..

[16]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[17]  Michael E Brier,et al.  Prediction of delayed renal allograft function using an artificial neural network. , 2003, Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association.

[18]  Ulrich Frei,et al.  Evaluation of the Eurotransplant Senior Program. The Results of the First Year , 2002, American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons.

[19]  Vladimir B. Bajic,et al.  Use of Artificial Neural Networks in Improving Renal Transplantation Outcomes , 2002 .

[20]  A. van der Lelij,et al.  Influence of HLA-A, HLA-B, and HLA-DR matching on rejection of random corneal grafts using corneal tissue for retrospective DNA HLA typing , 2001, The British journal of ophthalmology.

[21]  A. Levey,et al.  A More Accurate Method To Estimate Glomerular Filtration Rate from Serum Creatinine: A New Prediction Equation , 1999, Annals of Internal Medicine.

[22]  D. Gjertson,et al.  Prolonged hypertension (> 10 years) is a significant risk factor in older cadaver donor renal transplants. , 1999, Transplantation proceedings.

[23]  D. Shoskes,et al.  Prediction of early graft function in renal transplantation using a computer neural network. , 1998, Transplantation Proceedings.

[24]  Marko Robnik-Sikonja,et al.  An adaptation of Relief for attribute estimation in regression , 1997, ICML.

[25]  Edward C. Uberbacher,et al.  Use of neural networks for prediction of graft failure following liver transplantation , 1995, Proceedings Eighth IEEE Symposium on Computer-Based Medical Systems.

[26]  Bernhard E. Boser,et al.  A training algorithm for optimal margin classifiers , 1992, COLT '92.

[27]  J. H. Steiger.  Tests for comparing elements of a correlation matrix. , 1980 .

[28]  J. Blainey,et al.  Agonal phase, ischaemic times, and renal vascular abnormalities and outcome of cadaver kidney transplants. , 1975, British medical journal.

[29]  V. Vapnik.  Pattern recognition using generalized portrait method , 1963 .