This paper, first analysis the performance of image segmentation techniques; K-mean clustering algorithm and region growing for cyst area extraction from liver images, then enhances the performance of K-mean by post-processing. The K-mean algorithm makes the clusters effectively. But it could not separate out the desired cluster (cyst) from the image. So, to enhance its performance for cyst region extraction, morphological opening-by-reconstruction is applied on the output of K-mean clustering algorithm. The results are presented both qualitatively and quantitatively, which demonstrate the superiority of enhanced K-mean as compared to standard K-mean and region growing algorithm.

1. Chien-Cheng Lee, S. Chen and Y. Chiang, 2006. Automatic Liver diseases Diagnosis for CT Images using Kernel based classifiers in IEEE world Automation Congress, pp. 1-6.
2. Poonguzhali, B. Deepalakshmi and G. Ravindran, 2007 Optimal Feature Selection and Automatic Classification of Abnormal Masses in Ultrasound Liver Images in IEEE-ICSCN, pp. 503-506.
3. Rafael C. Gonzalez and Richard E. Woods, 2009. Digital Image Processing, Prentice-Hall.
4. Linda G. Shapiro and George C. Stockman (2001): Computer Vision, pp 279-325, New Jersey, Prentice-Hall, ISBN 0-13-030796-3.
5. C. Bezdek, L. O. Hall and L P. Clarke, 1993. Review of MR image segmentation techniques using pattern recognition in Med. Phys., vol. 20, Issue 4, pp. 1033-1048.
6. Ashish Thakur and Radhey Shyam Anand, 2010. A Local Statistics Based region Growing Segmentation Method for ultrasound Medical Images in Jr. of World Academy of Science, Engg. and Technology, pp. 321-326.
7. Madabhushi, A. and Mtaxas, D. N., 2003. Combining low-,high-level and empirical domain knowledge for automated segmentation of ultrasonic breast lesions in IEEE transactions on medical imaging, VOL. 22, NO. 2.
8. Eslami, A.; Kasaei, S.; Jahed, M., 2004. Radial multiscale cyst segmentation in ultrasound images of kidney in 4th IEEE International Symposium on Signal Processing and Information Technology, pp. 42-45.
9. Pavlidis, T., and Liow, Y. T., 1990. Integrating region growing and edge detection”, IEEE Transaction on Pattern Analysis and Machine Intelligence, 12, pp. 225-231.
10. XU D, GUO S. 2009. Research Progress of Liver CT Image Segmentation Techniques in Chinese Medical Equipment Journal•Vol．30 No．3.
11. Yi X., Zhong L. and Lin J., 2010. Liver CT Image Segmentation by Local Entropy Method in International Conf. on computer application and modelling, pp. VIII-591 –VIII-594.
12. Andrew Moore: K-means and Hierarchical Clustering - Tutorial Slides at http://www.cs.cmu.edu/~awm/tutorials/kmeans.html
13. Brian T. Luke: K-Means Clustering at http://fconyx.ncifcrf.gov/~lukeb/kmeans.html
14. Tariq R.: Clustering at http://www.cs.bris.ac.uk/home/tr1690/documentation/fuzzy_clustering_initial_report.
15. Image Processing Toolbox, available at http://www.caspur.it/risorse/softappl/doc/matlab_help/toolbox/images/morph11.html#34328.
16. L. Kaur, R.C. Chauhan, and S. C. Saxena, 2006. Wavelet based Compression of Medical Ultrasound images using Vector Quantization in Taylor and Francis JMET, pp. 128-133, vol. 30, no. 3.
17. S. Gupta, L. Kaur, R.C. Chauhan, and S. C. Saxena, 2007. A versatile Technique for visual enhancement of medical ultrasound images in Elsevier Jr. of Digital Signal Processing, 17, pp. 542-560.

Image segmentation region of interest k-mean clustering region growing k-mean clustering region growing