Detection of brain tumor in magnetic resonance images

Brain tumor is one of the serious diseases that have caused death to many people in recent years. The complex structure of the brain and its main function associated with the central nervous system and its main role in controlling most of the functions of the body makes detection of tumor a challenging task. Many techniques have been presented in the medical field in order to detect brain tumor from MRI images. In this thesis, a system is designed to detect the tumor in brain Magnetic Resonance Imaging (MRI) using a seed-based region growing algorithm (SBRG) in the segmentation process. Then after segmenting the region of interest (ROI) the texture features are extracted from the image and used to classify tumor and non-tumor tissues by the implementation of the Artificial Neural Network (ANN) classifier. The accuracy obtained by this system is 99.88%.