Skull Stripping of Mri Head Scans Based on Chan-vese Active Contour Model

Whole brain segmentation referred as skull stripping, it is an important process in neuriomage analysis. Automatic segmentation of brain tissues from magnetic resonance images (MRI) remains a challenging task due to variation in shape and size, use of different pulse sequences, overlapping signal intensities and imaging artifacts. Level sets and active contour methods have tremendous potential in the area of image segmentation. In this paper, we propose a new skull stripping algorithm for magnetic resonance images (MRI) of human head scans based on Chan-Vese active contour method. This is a fully automatic method for segmenting the brain from other non-brain tissues in T1, T2 and PD weighted MR images. The proposed method consists of two major processes. First we extracted the brain in the middle slice and then the brains in the remaining slices are extracted. In this method the binary form of the brain image is processed first to find the rough brain. The initial contour is drawn inside the rough brain portion to propagate the active contour. In our method the process of extracting the brain in the remaining slices were simplified by using the geometric similarities of the adjacent slice. The proposed method extracts the brain accurately in T1, T2 and PD weighted images. The result of the proposed method is compared with the standard manual skull stripping gold standard images and produced significant result. The experimental results indicate that the proposed method accurately extracted the brain which is comparable to that of BET and BSE using the IBSR and WBA datasets.