T. JEMAL, N. FIKADU, B.S. GOSHU^#

Department of Physics, Dire Dawa University, Ethiopia, 

The Medical image processing is the most emergent field these days and processing of images scanned through the magnetic resonance images (MRI) are the part of it. The main Purpose of this study is to extract and detect brain tumor from magnetic resonance scanned images. This study was using MATLAB by developing graphical user interface to extract the brain tumor from MRI scan image. Raw image converted to gray, some noise removal, morphological operation, segmentation using the threshold and region growing was done during coding. All 56 selected brain tumors were tested. About 94.64 % (53) were detected correctly and the rest of 5.36 % (3) were partially detected, since the images have complex nature in their composition. Those cystic (fluid filled) appear dark in T-W MRI which was difficult to detect as tumor in this algorithm. Image feature extraction from brain MRI using the Graphic User Interface (GUI) is detecting the tumor and its geometrical feature extraction.

Key words: Brain tumor, MRI, threshold segmentation, MATLAB, GUI.

Corresponding author’s e-mail: belaysitotaw@gmail.com