ANFIS models for prognostic and survival rate analysis “nasopharyngeal carcinoma”.

Fuzzy modeling and identification methodologies have been successfully used in a number of real-world applications. The Takagi-Sugeno model has often been employed in the modeling and identification of nonlinear technical processes from data. In this context we propose a new fuzzy inference system designed specifically to predict the survival rate in a given medical data. In this study we are concerned with NPC because it is one of the most common cancers in Malaysia. Two training methods were used namely back propagation and a hybrid method to train the FIS model. These two models were performed to evaluate the predictive accuracy, and the results were found to be satisfactory.