Segmentation of Microscopic Images for Counting Leukocytes

Counting of leukocytes in peripheral blood is commonly used in basic clinical diagnosis. The reliable approach to count leukocytes depends on morphological assessment of cellular materials under a light microscopy. Microscopic evaluation by medical technologists is a time-consuming and stuffy job. It is of interest to apply computer-aided analysis with image processing and pattern recognition, instead of human to count leukocytes. In this paper, a novel segmentation method is developed to count leukocytes by incorporating textural information. The proposed method integrates a non-decimated (or called shift invariant) complex wavelet transform into watershed segmentation, and uses information obtained by adaptive threshold segmentation to refine the result of watershed segmentation. Experiments demonstrate that the proposed method can obtain satisfactory results in comparison with the judgments of medical technologists.

[1]  Jean-Philippe Thiran,et al.  A queue-based region growing algorithm for accurate segmentation of multi-dimensional digital images , 1997, Signal Process..

[2]  Dorin Comaniciu,et al.  Image-guided decision support system for pathology , 1999, Machine Vision and Applications.

[3]  Richard Beare.  A locally constrained watershed transform , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  James M. Keller,et al.  Snakes on the Watershed , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[5]  Pedro Gómez Vilda,et al.  An improved watershed algorithm based on efficient computation of shortest paths , 2007, Pattern Recognit..

[6]  Sokratis Makrogiannis,et al.  A region dissimilarity relation that combines feature-space and spatial information for color image segmentation , 2005, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[7]  Peter Meer,et al.  Unsupervised segmentation based on robust estimation and color active contour models , 2005, IEEE Transactions on Information Technology in Biomedicine.

[8]  Hubert Cardot,et al.  Cooperation of color pixel classification schemes and color watershed: a study for microscopic images , 2002, IEEE Trans. Image Process..

[9]  Ron Kikinis,et al.  Improved watershed transform for medical image segmentation using prior information , 2004, IEEE Transactions on Medical Imaging.

[10]  Jing Li Wang,et al.  Color image segmentation: advances and prospects , 2001, Pattern Recognit..

[11]  S. Napel,et al.  Medical image segmentation using analysis of isolable-contour maps , 2000, IEEE Transactions on Medical Imaging.

[12]  Luciano da Fontoura Costa,et al.  A texture approach to leukocyte recognition , 2004, Real Time Imaging.

[13]  Luc Vincent,et al.  Watersheds in Digital Spaces: An Efficient Algorithm Based on Immersion Simulations , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Cedric Nishan Canagarajah,et al.  Image segmentation using a texture gradient based watershed transform , 2003, IEEE Trans. Image Process..

[15]  J. Macgregor,et al.  Image texture analysis: methods and comparisons , 2004 .

[16]  Dorin Comaniciu,et al.  Computer-assisted discrimination among malignant lymphomas and leukemia using immunophenotyping, intelligent image repositories, and telemicroscopy , 2000, IEEE Transactions on Information Technology in Biomedicine.