Fuzzy Region-Scalable Fitting Model for Glioma Brain Tumor Segmentation

Brain tumor segmentation in Magnetic Resonance images is one of the challenges in image processing because MR images have intensity inhomogeneity. In this paper, a new regionbased active contour model is proposed for Glioma Brain Tumor Segmentation in MR images. This model is based on the combination of fuzzy c-means clustering and regionscalable fitting energy model which can automatically segment images with intensity inhomogeneity. BraTS 2017 dataset was used for glioma brain tumor segmentation. Experimental results on inhomogeneous brain MR images indicate the superiority of the proposed model over the other state-of-the-art segmentation methods in the accuracy of segmentation based on the Dice similarity coefficient.

Keywords:

Brain tumor segmentation , Regin- Scalable Fitting , fuzzy c-means clustering , level set method.