Brain disorders are the most common fatality in the current scenario of health care society. An increasing incidence of brain disorders has been reported for the last three decades. Although the survival rate continues to remain poor, there are several reports suggesting an increasing trend in incidence rates. The bulk of this increase is reportedly due to rising incidence rates among the elderly and children. Hence, accurate detection of the type of the brain abnormality is highly essential for treatment planning which can minimize the fatal results. Hence this work is motivated to develop the techniques for accurate segmentation of neonates and premature infants.