A Novel Approach for Brain Tumor Detection Using Anisotropic Filtering and SVM Classifier from MRI

Abstract

Brain tumor is a dangerous disease and its early detection is very important to save life. The tumor region can be
detected by segmentation of brain Magnetic Resonance Image (MRI). With the help of radiologic evaluations the
suspected brain tumor location and size of tumor can be determined. The report of this decision is very important
for further diagnosis and treatment. The early and correct diagnosis of brain tumors plays an important role. In
order to extract tumor from MRI images of brain different image segmentation techniques are used. For cancer
diagnosis the brain tumors segmentation can be done manually from MRI, which gives the poor level of accuracy
and identification. So, The classification of abnormalities manually is not predictable and it is a time consuming
task for physician. Hence the automatic segmentation of brain tumors are important research area in present day
scenario. Several automated segmentation algorithms have been proposed. But still segmentation of MRI brain
image remains as a challenging problem due to its complexity and there is no standard algorithm that can produce
satisfactory results. The aim of this research work is to propose and implement an efficient system for tumor
detection and classification. The different steps involved in this work are image preprocessing for noise removal,
feature extraction, segmentation and classification. Proposed work preprocessed the MRI brain image using
anisotropic diffusion filter for noise removal, By applying the fast bounding box (FBB) algorithm, the tumor area
is displayed on the MRI image with a bounding box and the central part, and SVM classifier for segmentation and
morphological operations for separating the affected area from normal one.