A Combination of Bias-Field Corrected Fuzzy C-Means and Level Set Approach for Brain MRI Image Segmentation

Entire brain consists of several tissues specifically gray matter (GM), white matter (WM) and cerebrospinal fluid CSF. From brain image it is troublesome to delineate these tissue regions exclusively since these regions are not well defined by sharp boundaries. In present paper a combination of approaches namely bias-field corrected fuzzy C-means and level set segmentation are presented for brain MRI image segmentation into white and gray matter. Initially image is processed against bias field, then this bias corrected image is further processed using fuzzy C-means segmentation and after that level set segmentation. The results obtained using combination of bias field correction, FCM and level sets based segmentation have been proved better on comparing with individual FCM and level set segmentation methods. The combination approach was tested on 60 images and the results were quite good.