A Mixed Model Based on Watershed and Active Contour Algorithms for Brain Tumor Segmentation

Abstract

Segmentation of anatomical regions is the fundamental problem in medical image analysis. The watershed algorithm is used for the segmentation of anatomical regions and it is computationally simple. The active contour algorithm is used to extract the tumor region in the segmented image, but it suffers from computational complexity and it is insensitive to noise. The proposed method combines watershed algorithm with active contour algorithm to reduce the computational complexity and to improve the insensitiveness to noise. The real brain MR image is first segmented using watershed segmentation and then edges produced will be the initial contour of active contour algorithm. The proposed method is superior in terms of capture range, concave tumor region extraction, segmentation accuracy and sensitivity to noise.