Abstract: A manual identification is labour intensive and tedious due to the large amount of medical data to be processed and the presence of small lesions. Automated tumour segmentation from magnetic resonance imagery (MRI) plays a significant role in cancer research and clinical practice. However, tumour segmentation is an extremely challenging task: clinicians believe that a gamut of prior domain knowledge and clinical data should be used, along with the MR image. In this paper a novel approach for segmenting the structure of malignant brain tumour is proposed. The method is general enough to segment different types of abnormalities.

Keywords: Brain Tumour, Region growing, MRI, Segmentation, Texture analysis, Edge detection etc.