Brain tumor is one of the major causes for the increase in mortality among the children and adults. In fact, tumor is a mass of tissue that grows-out of control from the normal forces which regulates growth. Therefore, identification of this in advance helps in the recovery of patients through suggestive and corrective treatments. To identify these brain diseases, it is essential to segment the brain tissues which consist of mainly three parts such as Gray Matter (GM), White Matter (WM) and Cerebro spinal fluid (CSF). There are many segmentation techniques available based on parametric and non-parametric models. Among these models, segmentation of medical images based on parametric technique is more accurate. In model based segmentation, entire image is viewed as a collection of image region. Finite mixture models are utilized to characterize the pixel intensities inside the images. Hence, Finite Skew Gaussian Mixture model is used to carry out the segmentation process, the initial parameters are obtaining by using clustering algorithms and the updated equations are obtained by deriving the equations using EM algorithm.