| Abstract |
: |
Brain tumor occurs due the uncontrolled growth of brain tissues. The detection of size, shape, type, location and orientation of the brain abnormality is highly essential for planning effective treatment. Magnetic resonance imaging (MRI) is a traditional and most commonly used technique for detecting brain tumors, cancer, multiple sclerosis and other abnormalities. Nowadays Computer Aided Diagnosis (CAD) systems are commonly used for systematic and explicit detection of brain abnormalities. Image segmentation is an effortful and tedious step in CAD. Image segmentation is used to subdivide an image, and is an important step in a CAD system. The representation of the image is changed and a meaningful image is obtained, which can be used for better analysis. The effectiveness of abnormality detection depends on the accuracy and robustness of segmentation algorithm. Segmentation techniques with different level of sensitivity, efficiency, and accuracy have been developed. In this paper we summarize, and discuss the advantages, capabilities and drawbacks of the most commonly used MRI segmentation methods. |
