Hepatoma cells recognition based on matrix absolute gray relational degree of B-mode

Abstract This paper proposed a novel hepatoma cells recognition method, which used Matrix Absolute Gray Relational Degree of B-Mode (MAGRD-B) to construct feature description of liver cells for distinguishing the cancerous cells from the normal ones. MAGRD-B proposed by us can be considered as a measure of the similarity and the proximity between different cells. And MAGRD-B can work without extracting the exact contours of cells. k-Nearest neighbor (k-NN) algorithm was utilized for determining whether liver cells are healthy or not. Experimental results are obtained on the clinical liver pathological images from different patients, which indicated that MAGRD-B can be used as a commendable feature to recognize hepatoma cells with better recognition accuracy.

[1]  Christophoros Nikou,et al.  Combining shape, texture and intensity features for cell nuclei extraction in Pap smear images , 2011, Pattern Recognit. Lett..

[2]  Kotagiri Ramamohanarao,et al.  Breast-Cancer identification using HMM-fuzzy approach , 2010, Comput. Biol. Medicine.

[3]  John Skilling,et al.  Maximum entropy method in image processing , 1984 .

[4]  Stanislaw Osowski,et al.  Recognition of Colon Cells Using Ensemble of Classifiers , 2007, 2007 International Joint Conference on Neural Networks.

[5]  Masatoshi Kitaoka,et al.  Circular Data Analysis of Periodic Time Series Data Using Grey Theory , 2007 .

[6]  Pranab Dey,et al.  Fractal dimensions of breast lesions on cytology smears , 2003, Diagnostic cytopathology.

[7]  D. Cavouras,et al.  An image-analysis system based on support vector machines for automatic grade diagnosis of brain-tumour astrocytomas in clinical routine , 2005, Medical informatics and the Internet in medicine.

[8]  Po-Whei Huang,et al.  Effective segmentation and classification for HCC biopsy images , 2010, Pattern Recognit..

[9]  H. T. Nagle,et al.  Feature extraction by genetic algorithms for neural networks in breast cancer classification , 1995, Proceedings of 17th International Conference of the Engineering in Medicine and Biology Society.

[10]  Wang Liyun,et al.  Spermatogonium image recognition using Zernike moments. , 2009, Computer methods and programs in biomedicine.

[11]  Bayan S. Sharif,et al.  Fractal analysis in the detection of colonic cancer images , 2002, IEEE Transactions on Information Technology in Biomedicine.

[12]  Ahmed Bouridane,et al.  An evolutionary snake algorithm for the segmentation of nuclei in histopathological images , 2004, 2004 International Conference on Image Processing, 2004. ICIP '04..

[13]  Josef Smolle,et al.  Tissue counter analysis of benign common nevi and malignant melanoma , 2003, Int. J. Medical Informatics.

[14]  Chen Guangsheng,et al.  Wood cell recognition using geodesic active contour and principal component analysis , 2013 .

[15]  S. B. Duarte,et al.  Computational cancer cells identification by fractal dimension analysis , 2009, Comput. Phys. Commun..

[16]  Cao Xiao-lin.  Matrix absolute grey relation degree and its significance , 2009 .

[17]  Stanislaw Osowski,et al.  Recognition and classification of colon cells applying the ensemble of classifiers , 2009, Comput. Biol. Medicine.

[18]  Panagiota Spyridonos,et al.  An automatic classification system of urine bladder tumors employing morphological and textural nuclear features , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[19]  Chie-Bein Chen,et al.  Concurrent maximization of process tolerances using grey theory , 2000 .

[20]  Z. Budimlija,et al.  Fractal dimension of hepatocytes' nuclei in normal liver vs hepatocellular carcinoma (HCC) in human subjects - preliminary results , 2000 .

[21]  Peter E. Hart,et al.  Nearest neighbor pattern classification , 1967, IEEE Trans. Inf. Theory.