Русский
Prediction of renal transplantation outcome using artificial neural networks and investigating important risk factors
DOI: https://dx.doi.org/10.18565/urology.2023.4.82-89
Abolfazl Zanghaei, Zohreh Rostami, Ali Ameri, Mahmood Salesi, Ahmad Akhlaghi, Ahmad Shalbaf, Hassan Doosti
1) Department of Biomedical Engineering and Biophysics, School of Medicine, Shahid Beheshti University of Medical Sciences, Tehran, Iran; 2) Research Center of Nephrology, School of Medicine, Baqiyatallah University of Medical Sciences, Tehran, Iran; 3) Chemical Injuries Research Center, Systems Biology and Poisonings Institute, Baqiyatallah University of Medical Sciences, Tehran, Iran; 4) Department of Health Information Management, School of Health Management & Information Sciences, Iran University of Medical Sciences, Tehran, Iran; 5) School of Mathematical and Physical Sciences, Macquarie University, Sydney, Australia
Background. Renal Transplantation is the final choice for some patients with ESRD (End-Stage Renal Disease), but some transplantations suffer from acute or chronic rejection, so it’s very important to predict the outcome of transplantation.
Methods. The dataset was extracted from records of 4572 patients with kidney transplantations. We applied an Artificial Neural Network (ANN) model to predict transplantation outcome. Moreover, novel features have been explored which enhanced the prediction performance.
Results. The results show that the well configured neural networks can predict renal transplant outcome with a sensitivity and specificity of higher than 86%. The results show creatinine is the most important risk factor that affects the renal transplantation outcome.
Conclusion. The designed neural networks can properly predict the transplantation outcome with the accuracy of 86%. Recipient creatinine is the most important variable in the prediction of the renal outcome.
Methods. The dataset was extracted from records of 4572 patients with kidney transplantations. We applied an Artificial Neural Network (ANN) model to predict transplantation outcome. Moreover, novel features have been explored which enhanced the prediction performance.
Results. The results show that the well configured neural networks can predict renal transplant outcome with a sensitivity and specificity of higher than 86%. The results show creatinine is the most important risk factor that affects the renal transplantation outcome.
Conclusion. The designed neural networks can properly predict the transplantation outcome with the accuracy of 86%. Recipient creatinine is the most important variable in the prediction of the renal outcome.
About the Autors
Corresponding author: Ali Ameri – Department of Biomedical Engineering and Biophysics, School of medicine, Shahid Beheshti University of Medical Sciences, Tehran, Iran. Email: aliameri86@gmail.com
Similar Articles
