Segmentation of images is an important part of computer based medical applications for diagnosis and analysis of anatomical data. With rapid advances in medical imaging modalities and various visualization techniques, computer based diagnosis is fast becoming a reality. These computer based tools allow scientists and physicians to understand and diagnose anatomical structures by virtually interacting with them. Image segmentation plays a critical role by facilitating automatic or semiautomatic extraction of the anatomical organ or region-of-interest. However, due to the specific and complex requirements of biomedical image segmentations, general image segmentation algorithms are either not applicable or need to be revised for accomplishing this image analysis task. This paper presents the analysis of implementation of the Watershed Algorithm for image segmentation for 2D medical image segmentation. The algorithm works on structural information and manipulation. While keeping the basic principal intact in mind, some filtering steps are incorporated to improve the image and the object location identification before applying the core algorithm. The algorithm is tested for multiple images and also multiple times to test its repeatability. A statistical analysis of the algorithm on the basis of the factors like segmentation time, accuracy, repeatability and generality and applicability is conducted on a set of input MRI and CT Scan images and the results are offered through this paper.