A New Level Set Method for Biomedical Image Segmentation

This paper presents a new biomedical image segmentation method that applies an edge-based level set method. According to low contrast in biomedical images, we mainly make focus on introducing the Laplace operator in external energy of level set method for accurately detecting object edge. A preliminary evaluation of the proposed method mainly performs on gallstone detection and extraction, mammographic image segmentation, iris inner location and polysaccharides extraction. Finally, the comparison experimental results demonstrate that our proposed approach potentially performs better than the representative level set method for biomedical image segmentation in terms of sensitivity, accuracy and specificity, with same initial contours.