Unsupervised Brain Tumor Segmentation from Magnetic Resonance Images

The segmentation of magnetic resonance imaging (MRI) is an essential step for many applications in medical fields. The detection of the tumor region and the precise recognition of the size and location of the tumor play an important role in the diagnosis. This is a very difficult task because of the complex structure of the brain and the complexity of tumor size. Several approaches have been proposed to help a better visualization of the appearance and severity of the tumor concerned. In this paper, we compare the performance of five fuzzy segmentation methods and we apply them on medical imaging on the one hand to identify the tumor area and on the other hand to determine the algorithm that gives a better calculation time. The comparison is based on the segmentation of a database of three MRI images of the brain.