Mathematical morphology aided enhancement and segmentation of T2-weighted brain MRI images

Abstract

Segmentation of the brain MRI into its constituent White Matter (WM), Gray Matter (GM) and Cerebrospinal Fluid (CSF) is a vital task for the diagnosis of various neurological diseases. In this work, mathematical morphology has been employed for contrast enhancement of the brain T2-wighted MRI, followed by segmentation with the help of Fuzzy C-means (FCM) clustering algorithm. The proposed method has resulted in better results, both qualitatively and quantitatively in comparison to the conventional method where FCM is directly applied to the raw MR images.