Paper Title
Detection of Brain Tumor Using Fast and Robustfuzzy C-Means Clustering Along With Supportvector Machine for Classification

Abstract
Abstract - Fuzzy c-means clustering (FCM) algorithm is sensitive to noise, therefore local spatial information is generally introduced to an objective function to improve the robustness of the FCM algorithm for image segmentation. Although, the addition of local spatial information gives way to a high and precise computational complexity, it causes an iterative calculation of the distance from one pixel to another amongst local spatial neighbors and clustering centers. To find a solution to this problem, an enhanced modified version of Fast and Robust Fuzzy C-Means Clustering (FRFCM) using morphological reconstruction and membership filtering has been proposed. The benefit of this improved algorithm is that it indeed overcomes the constraints and limitations of traditional FCM technique. In this paper, the proposed methodology has been used for brain tumor detection from MRI images, and further if the presence of a tumor is confirmed then it evaluates its stage as either Benign, or Malignant. This paper is focused towards the design of a more accurate, precise and optimal way for the detection and classification of tumor from brain MRI scans using a unique combination of morphological reconstruction, memberships filtering, fast and robust fuzzy C-Means for detection and further classification using Principal component analysis of the image on several parameters and Support vector machine classifiers. It has been experimentally demonstrated that the proposed model has a more prominent and higher precision than other existing state-of-the-art algorithms for characterizing the tumor stage with an accuracy for discovery of Malignant tumor as high as99.02% and for Benign tumor, an accuracy of 99.67%. Keywords - Fuzzy c-means, Clustering, FRFCM, Brain Tumor, Support Vector Machine, Morphological Reconstruction