3D SEGMENTATION OF GLIOMA FROM BRAIN MR IMAGES USING SEEDED REGION GROWING AND FUZZY C-MEANS CLUSTERING

Tejus Thirumeni, Renu John, Sikandar Shaikh

Abstract: This paper reports 3-D segmentation of glioma from brain MR images. We discuss two algorithms for brain MR image segmentation. The images used are axial MR images of the human brain. The images show a glioma. The objective is to segment the tumor and edema surrounding it from the images. Initially the images are pre-processed by contrast adjustment. Segmentation is performed by two algorithms: seeded region growing and fuzzy c-means clustering. After the images are segmented, the volumes of the segmented regions are measured. The segmentation has been performed in MATLAB. Finally the results are rendered in 3D in AMIRA.

Keywords: MRI, glioma, contrast adjustment, seeded region growing, fuzzy c-means clustering

DOI: https://doi.org/10.15623/ijret.2015.0424014