A Novel Classification Algorithm based on Contextual Information using FCM Classifier for Brain Tumor Diagnosis using MR Images

The identification and localization of a brain tumor is a complex work. The proposed method allows to assess the anatomical location, morphology and extent of the tumor with high spatial resolution. A modified Fuzzy C-Means (FCM) Classifier is introduced to identify and locate the tumor effectively with minimal duration. The brass tacks of the classifier is carried out by incurring the contextual information from the brain tissues. The clustering is used to classify the tissues expeditiously and the standing of specifying the cluster size is user customizable. Due to the intricateness of the human brain the tumor may overlaps with the healthy brain tissues in Magnetic Resonance (MR) images. FCM works on an ingeminating fashion to discriminate the tumor from normal tissues.