Fuzzy based uncertainty modeling of Cancer Diagnosis System

Recent advances in the field of artificial intelligence and machine learning have led to the emergence of expert systems for medical applications. In this paper a machine learning approach is implemented to diagnose cancer. The method developed for the diagnosis of cancer is to detect, whether it is benign or malignant using the artificial neural network. The network is trained using back-propagation algorithm. In the further test the membership values are calculated using fuzzy-c-means algorithm. They depict the possibility for a benign to turn into malignant. Then a neuro-fuzzy system is developed which is based on the mamdani model. The rules are set up using the pruned and un-pruned decision tree to identify the magnitude of cancer.