Region based techniques for segmentation of volumetric histo-pathological images

In this article we have presented the application of three region based seg mentation techniques namely, seeded volume growing, constrained erosion-dil ation techniques and 3-D watershed algorithm. The algorithms are suitably e xtended to apply on 3-D histo-pathological images. Suitable modifications a nd extension for each algorithm is done to obtain better segmentation. A qu antitative as well as qualitative comparison of the three methods is presen ted. Modifications to these algorithms for obtaining better results are dis cussed. The modifications include, (1) design of adaptive similarity measur es to control the seeded volume growing and (2) rule-based merging of the o ver-segmented cells in the case of the 3-D watershed algorithm. Some result s and quantitative study is also presented. (C) 2000 Elsevier Science Irela nd Ltd. All rights reserved.