Analysis of Graph Cut Technique for Medical Image Segmentation

Abstract

Segmentation plays an important role in image analysis as it is used to identify and differentiate foreground and background regions. Image seg mentation in brain MRI analysis performs several roles like extraction of abnormal region for better diagnosis of the disease aiding in the therapy plan ning. Various brain tumors comprise diverse properties like their shapes, intensity distribution and location, hence reducing the possibility of developing a single general algorithm. In this paper authors have illustrated two methods for performing extraction which includes histogram thresholding and centroid based graph cut segmentation. On the basis of their potential, advantages and limita tion comparison is made, that emphasize better performance of centroid based graph cut segmentation method. To measure the performance some quality parameters are evaluated. This paper also solves the problem of initial seed selection by using graph cut segmentation technique.