Brain Tumor Detection Using Anisotropic Filtering, SVM Classifier and Morphological Operation from MR Images

Abstract

Tumor is a pre-stage of cancer which has become a serious problem in this era. Researchers are trying to develop methods and treatments to round it. Brain tumor is an exceptional cell enhancement in brain tissue and may not always be seen in imaging tricks. Magnetic Resonance Imaging (MRI) is a technique which is applied to display the detailed image of the attacked brain location. The medical imaging trick plays a significant behavior in identification of the disease. In this paper, the brain MRI image is chosen to investigate and a method is targeted for more clear view of the location attacked by tumor. An MRI abnormal brain images as input in the introduced method, Anisotropic filtering for noise removal, SVM classifier for segmentation and morphological operations for separating the affected area from normal one are the key stages if the presented method. Attaining clear MRI images of the brain are the base of this method. The classification of the intensities of the pixels on the filtered image identifies the tumor. Experimental result showed that the SVM has obtained 83% accuracy in segmentation. Finally, the segmented region of the tumor is put on the original image for a distinct identification.

Ähnliche Arbeiten