Support vector machine based prognostic analysis of renal transplantations

Chronic renal disease is a common problem faced by people. Renal transplantation is the only treatment of choice for patients with end-stage renal disease. Prediction of a graft transplant outcome and the survival analysis with a high accuracy level is a challenging as well as very demanding task. The solution for this challenge lies in data mining and machine learning. An intuitive understanding of the post transplantation interaction mechanisms involving graft and host is intricate and on account of this prognosis of planned organ transplantation outcomes is an involved problem. Prediction methods based on donor and recipient data are indespensible for improved prognosis of graft outcomes. In our study a new method is developed to make the prediction process more efficient with improved donor recipient matching. Clinical information was gathered prospectively for 221 patients who underwent kidney transplantation. In this study we used two levels of classification: clinical and non clinical features separately implemented in SVM. The Cox regression analysis was done to find the most critical determent features.