Conference Paper

Level Set Segmentation of Cellular Images Based on Topological Dependence

DOI: 10.1007/978-3-540-89639-5_52 Conference: Advances in Visual Computing, 4th International Symposium, ISVC 2008, Las Vegas, NV, USA, December 1-3, 2008. Proceedings, Part I
Source: DBLP


Segmentation of cellular images presents a challenging task for computer vision, especially when the cells of irregular shapes
clump together. Level set methods can segment cells with irregular shapes when signal-to-noise ratio is low, however they
could not effectively segment cells that are clumping together. We perform topological analysis on the zero level sets to
enable effective segmentation of clumped cells. Geometrical shapes and intensities are important information for segmentation
of cells. We assimilated them in our approach and hence we are able to gain from the advantages of level sets while circumventing
its shortcoming. Validation on a data set of 4916 neural cells shows that our method is 93.3 ±0.6% accurate.

Download full-text


Available from: Weimiao Yu, May 26, 2014
  • Source
    • "Timings for all figures are presented in appendix B. In our future work, we would like to extend our statistical physics approach to include topological dependence [31]–[33] and the subspace Mumford–Shah model [9]–[11]. Eden clustering can be used to include a topological dependence constraint developed recently to segment crowded objects [31] [32]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Natural images are depicted in a computer as pixels on a square grid and neighboring pixels are generally highly correlated. This representation can be mapped naturally to a statistical physics framework on a square lattice. In this paper, we developed an effective use of statistical mechanics to solve the image segmentation problem, which is an outstanding problem in image processing. Our Monte Carlo method using several advanced techniques, including block-spin transformation, Eden clustering and simulated annealing, seeks the solution of the celebrated Mumford–Shah image segmentation model. In particular, the advantage of our method is prominent for the case of multiphase segmentation. Our results verify that statistical physics can be a very efficient approach for image processing.
    Full-text · Article · Jan 2011 · New Journal of Physics
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Automatic and quantitative measurement of neurites is a challenging task, while it is critical in many neurological studies. We developed a fully automatic method to trace and quantitatively measure the neurites. Our measurements are validated by comparison with the semi-automatic NeuronJ and commercial software HCA- Vision. The results demonstrate that the measurements of the three approaches have no significant difference. We also apply our approach for a biological study on neurite outgrowth and the measurements of four different conditions are presented.
    Full-text · Article ·
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Neural stem cells derived from both embryonic and adult brain can be cultured as neurospheres; a free floating 3-D aggregate of cells. Neurospheres represent a heterogenous mix of cells including neural stem and progenitor cells. In order to investigate the self-renewal, growth and differentiation of cells within neurospheres, it is crucial that individual nuclei are accurately identified using image segmentation. Hence effective segmentation algorithm is indispensible for microscopy based neural stem cell studies. In this paper, we present a seed finding approach in scale space to identify the center of nuclei in 3-D. Then we present a novel segmentation approach, called “Evolving Generalized Voronoi Diagram”, which uses the identified centers to segment nuclei in neurospheres. Comparison of our computational results to mannually annotated ground truth demonstrates that the proposed approach is an efficient and accurate segmentation approach for 3-D neurospheres.
    Full-text · Conference Paper · Nov 2009
Show more