Skull Stripping Magnetic Resonance Images Brain Images: Region Growing versus Mathematical Morphology

Abstract

Skull stripping is a major phase in MRI brain imaging applications and it refers to the removal of the brain’s non-cerebral tissues. The main problem in skull-stripping is the segmentation of the non-cerebral and the intracranial tissues due to their homogeneity intensities. Numerous techniques were applied in the studies of skull stripping, most common are region growing and mathematical morphology. This paper investigated the strength and weaknesses of these two methods on three types of MRI brain images. Unlike previous researches which normally tested on one type of MRI images only, this paper experimented on ninety samples of T1-weighted, T2-weighted and FLAIR MRI brain images. Qualitative evaluations showed that skull stripping using mathematical morphology outperformed region growing at an acceptance rate of 95.5%, whereas quantitative evaluation using Area Overlap, False Positive Rate and False Negative Rate produced of 96.2%, 2.2% and 1.6% respectively.