AUTOMATIC SEGMENTATION OF NUCLEI IN 3D MICROSCOPY IMAGES OF C.ELEGANS

摘要: Automatic segmentation of nuclei in 3D microscopy images is essential for many biological studies including high throughput analysis gene expression level, morphology, and phenotypes single cell level. The complexity variability the present difficulties to traditional image methods. In this paper, we a new method based on watershed algorithm segment such images. By using both intensity information geometry appropriately detected foreground mask, our robust fluctuation within at same time sensitive geometrical cues between nuclei. Besides, can automatically correct potential errors by several post-processing steps. We tested confocal C.elegans, an organism that has been widely used studies. Our results show accuracy despite non-uniform background, tightly clustered with different sizes shapes, fluctuated intensities, hollow-shaped staining patterns