Fuzzy Clustering and Deformable Model for Tumor Segmentation on MRI Brain Image: A Combined Approach

Deformable models are extensively used for medical image segmentation, particularly to locate tumor boundaries in brain tumor MRI images. Problems associated with initialization and poor convergence to boundary concavities, however, has limited their usefulness. As result of that they tend to be attracted towards wrong image features. In this paper, we propose a method that combine region based fuzzy clustering and deformable model for segmenting tumor region on MRI images. Region based fuzzy clustering is used for initial segmentation of tumor then result of this is used to provide initial contour for deformable model, which then determines the final contour for exact tumor boundary for final segmentation using gradient vector field as a external force field. The evaluation result with tumor MRI images shows that our method is more accurate and robust for brain tumor segmentation.