An Improved Approach for Breast Cancer Detection in Mammogram based on Watershed Segmentation

An accurate technique for breast tumor segmentation is a critical step for monitoring and quantifying breast cancer. The fully automated tumor segmentation in mammograms presents many challenges related to characteristics of an image. In this paper, a hybrid segmentation algorithm, which combines the watershed transform and level set techniques, is proposed. Since watershed segmentation is based on pixel density variation that is present in all mass tumors, it was fairly successful in locating tumors under all conditions. However it is very sensitive to small variations of the image magnitude and consequently the number of generated regions is undesirably large and the segmented boundaries are not smooth enough. Meanwhile Level set methods offer a powerful approach for the medical image segmentation since it can handle any of the cavities, concavities, convolution, splitting or merging. However, this method requires specifying initial curves and can only provide good results if these curves are placed near symmetrically with respect to the object boundary. In our proposed technique a watershed segmentation algorithm was developed to initially locate breast mass tumor candidates. In order to facilitate and improve the detection step, the segmentation results is treated as the initial localization of the desired contour, and used in the following level set method, which provides closed, smoothed and accurately localized contours or surfaces. Experimental results show the significant improvement of the final segmentation accuracy. General Terms Image Processing.