Integrating Spatial Fuzzy Clustering with Level Set Methods for Improving Medical Imaging Segmentation

The main purpose of medical image segmentation is to divide the image into anatomically different structures that makes possible the separation of desired parts, such as blood vessels and liver tumors, from their background. Low contrast and resolution make medical image segmentation complex. Moreover, noises and instrumental limitations of reconstruction algorithms and body movements of the patient adds to the complexity of the task. Appropriate initialization and optimal configuration need a high level of manual adjustments. On the other hand, appropriate initialization and optimal configuration of controlling parameters influence the performance of level set segmentation. This paper aimed at analyzing, implementing, and optimizing level set algorithm using fuzzy logic. The algorithm carried out initial segmentation using fuzzy clustering. Controlling parameters and level set initialization were estimated using fuzzy clustering. Finally, level set algorithm was exploited.