In this paper, a robust shape extraction method for the medical image application was developed. The method combines object region statistical information with the level set method. The new method is based on conditional independence of the gray-level intensities in the different regions. It is posed within a Bayesian framework of maximization of a posterior probability. The energy function is minimized by the level set method. The level set implementation of the contour evolution supports topology changes for object contour. We have presented some preliminary experimental results illustrating the flavor of this technique. The experimental results show that incorporating region statistical information into the level set framework, an accurate and robust segmentation can be achieved