Segmentation of microscopic nuclear images-A review

Accurate and reliable segmentation of nuclei and their compartments is an important step for the quantitative image analysis of nuclear malignancy. To evaluate and analyze the properties of nuclei, such as the shapes, size, membrane, chromatin appearence and textures, nuclear regions need to be extracted and separated from the image background. Accurate segmentation of nuclei is often difficult because of the heterogeneous cellular staining and nuclear overlapping. In this article, we present a general review of recent developments on segmentation of cell images. Reviews include algorithms of histogrambased thresholding, region-based region growing, interactive, parametric and color classification. Correspondence/Reprint request: Dr. Hai-Shan Wu, Department of Pathology, Box 1194, Mount Sinai School of Medicine, One Gustave L. Levy Place, New York, NY 10029, USA. E-mail: [email protected] Hai-Shan Wu et al. 2