Efficient segmentation of brain tumor images using fuzzy and svm classifier approach

Abstract

Medical image processing is a quickly developing and a focusing area in current days. Medical image techniques are used to detect and treatment of diseases. One such basic and life-imperiling disease is brain tumor which is an abnormal development of brain cells inside the brain. Identification of brain tumor is an exploration because of the difficulties in the structure of the brain. In this work, to enhance the performance and lessen the intricacy includes in the image segmentation process, it has explored Computer Tomography (CT) based brain tumor segmentation. CT images are most ordinarily utilized for detection of head wounds, tumors, and Skull break. In this research work, brain tumor database pictures are considered under the preprocessing method called adaptive median filter is applied to improve the clearness of the image. In preprocessing stage, noise and high-frequency artifact present in the images are evacuated. The median filter is a nonlinear digital filtering strategy, frequently utilized for noise reduction on an image or signal. Notwithstanding the preprocessing methodology, feature extraction strategies are actualized and after that, the classification procedures, for example, Adaptive Neuro-Fuzzy Inference System (ANFIS) and Support Vector Machine (SVM) classifier are applied on the image to categories the images into normal and abnormal. After the classification, the abnormal pictures are monitored and selected for segmentation process employing Fuzzy C-Means (FCM) clustering process along with the involved optimization strategies. newline