Automatic brain MR perfusion image segmentation using adaptive diffusion flow active contours based on Modified Fuzzy C Means

In this paper, we are interested to segment brain MR perfusion image using active contours or deformable models in order to assist in diagnosis. Traditional methods are often unable to perform adequately on these types of images which have poor contrast, high-level speckle noise and boundary gaps. For this purpose, we propose a Modified Fuzzy C Means method combined with the Adaptive Diffusion Flow model. The proposed method can provide significantly improved performance with an accurate segmentation. The performance of the algorithm has been tested on Brain MR Perfusion Image.