Identifying the Factors Affecting the Survival Rate of Kidney Transplant Patients in Isfahan Using Classification Techniques

Abstract

Purpose: 10% of the world's population suffers from chronic kidney disease and millions of deaths occur annually due to lack of access to appropriate treatment in the world. Kidney transplantation is associated with several problems. These problems, including kidney rejection, the consequences of surgery, drug poisoning, and infectious diseases can reduce the chances of survival of these patients. The science of classification has been proposed in recent years to reduce medical errors due to inexperience, reduce the workload of physicians and provide a suitable model for making better decisions. Materials and Methods: The data set includes information about patients for whom kidney transplantation was performed in Isfahan. The data set includes 2554 patients and 38 attributes. The techniques used in this study will include random forest, Principal Component Analysis (PCA), and Support Vector Machine (SVM). Results: Among the studied techniques, PCA technique in three classes out of four classes had better performance than other techniques. The syndrome has the highest recurrence among traits. Five attributes include syndrome, blood type, dialysis time, weight, and age. Conclusion: The results showed that the PCA method in the case of non-numerical data has a good performance in identifying attributes. Also, five attributes that affect the survival rate of kidney transplant patients were identified.