Human brain tumors detection using neutrosophic c-means clustering algorithm

Abstract

For the last several decades, detecting human brain tumors has evolved into one of the most difficult problems in the field of medical research. In the realm of medical image processing, the categorization of brain tumors is a difficult job to do. In this research, we offer a model for the detection of human brain tumors in magnetic resonance imaging (MRI) images that makes use of the template-depend neutrosophic c-means and is compared with the fuzzy C means method. This model is referred to as the NCM method. In this suggested method, well first of all, the pattern K-means method is used to initialize segmentation markedly through the ideal choice of a template, depending on the gray-level intensity of the image; besides which, the revised membership is calculated by the ranges from the closest centroid to cluster pieces of data by using neutrosophic C-means (NCM) method while it approaches its perfect outcomes; and at last, the NCM clustering method is used for sensing tumor positron emission tomography (PET) imaging The findings of the simulation reveal that the suggested method can produce improved identification of pathological and normal cells in the human brain despite a little separation in the intensity of the grey level.