Hybrid method for white matter separation in brain images using granular rough sets and fuzzy thresholding

Medical image segmentation is a complex and challenging task due to the intrinsic nature of the images. The brain has particularly complicated structure and its precise segmentation is very important for detecting tumors, edema, and necrotic tissues, in order to prescribe appropriate therapy. Recently, rough sets and fuzzy sets has proved its soundness and usefulness in many medical applications including image segmentation. This paper presents a hybrid method that combines the granular rough set approach for brain image segmentation and fuzzy thresholding for brain white matter separation and the results show the effectiveness of the method.