An Automatic Cell Counting Method for a Microscopic Tissue Image from Breast Cancer
This paper presents an automatic cell counting method for a microscopic tissue image from breast cancer. We perform color space changing from RGB to CIELab and anisotropic diffusion filtering for noise removal in the preprocessing stage. Subsequently, the segmentation algorithm based on local adaptive thresholding, morphological operations, and cell size considerations is performed. In order to obtain the more correct counting number of cancer cells, we further process the image containing attached cancer cells with marker-controlled watershed segmentation. Results from our automatic counting approach show a promising solution to the traditional manual analysis. That is, the counting number of cancer cells from the automatic approach is comparable to that from a specialist.
KeywordsQuantitative immunohistopathology Image segmentation Cancer cell images
Unable to display preview. Download preview PDF.
- 3.Zhao P, Mao K, Koh T, Tan P (2003) Automatic cell analysis for P53 immunohistochemistry in bladder inverted papilloma. IEEE EMBS Asian-Pacific Conference on Biomedical Engineering, 2003, pp 168–169.Google Scholar
- 4.Petushi S, Katsinis C, Coward C et al (2004) Automated identification of microstructures on histology slides, IEEE International Symposium on Biomedical Imaging: Macro to Nano vol. 1, 2004, pp 424–427.Google Scholar
- 7.Phukpattaranont P, Boonyaphiphat P (2006) Automatic classification of cancer cells in microscopic images: Preliminary results, The 2006 ITC-CSCC International Conference vol. 1, Chiang Mai, Thailand, 2006, pp 113–116Google Scholar
- 8.Phukpattaranont P, Boonyaphiphat P et al (2006) Segmentation of cancerous cell image using local adaptive thresholding and morphological operators, The 2nd Regional Conference on Artificial Life and Robotics, Songkhla, Thailand, 2006, pp 68–71Google Scholar