Epithelial Cell Layer Segmentation UsingGraph-cut and Its Application in TesticularTissue

Computerized image processing has provided us with valuabl e tools for analyzing histology images. However, histology images are complex, a nd the algorithm which is developed for a data set may not work for a new and unseen data s et. The preparation procedure of the tissue before imaging can significantly aff ect the resulting image. Even for the same staining method, factors like delayed fixation d egrade tissue and may alter the image quality. In environmental research, due to the dis tance between the site where the wild animals are caught and the laboratory, there is alwa ys a delay in fixation. In this paper we face the challenging problem of designing a method t hat works on data sets with different fixation delay and strongly varying quality. Here w suggest a segmentation method based on the structural information of epithelium ce ll layer in testicular tissue. The cell nuclei are detected using the fast radial symmetry fi l er. A graph is constructed on top of the epithelial cells. Graph-cut optimization meth od is used to cut the links between cells of different tubules. The algorithm is tested on five different groups of animals. Group one is fixated immediately, four groups were l ft at room temperature for 6, 18, 30 and 42 hours respectively, before fixation. The s uggested algorithm gives promising results for the whole data set.