Texture analysis is one of the primary ways of extracting relevant information from digital images. Analysis of digital mammograms is essential in distinguishing between normal tissue and tissues that are showing early signs of breast cancer. In this paper, we compute certain Haralick texture features (Angular Second Moment, Contrast, Correlation and Entropy) and compare the performance of simple distance-measure classifications with each of these features, as well as the mean of all four. The correlation feature and the mean of all four features shows better accuracy when applied on digital mammograms to classify them into normal tissues and cancerous tissues.

1. Timo Ojala, Matti Pietikainen and David Harwood, A Comparative Study Of Texture Measures With Classification Based On Feature Distributions, Pattern recognition 29. 1 (1996): 51-59
2. Robert M. Haralick, K. Shanmugam and ITS'HAK Dinstein. Textural features for Image Classification, IEEE Transactions on Systems, Man, and Cybernetics, 1973, SMC-3 (6): 610–621
3. Eizan Miyamotol and Thomas Merryman Jr, Fast Calculation of Haralick Texture Features, Technical Report, Carnegie Mellon University, found at: www. ece. cmu. edu/pueschel/teaching/18-799B-CMU-spring05/ material/eizan-tad. pdf
4. M. Zikos, E. Kaldoudi, S. C. Orphanoudakis. DIPE: A Distributed Environment for Medical Image Processing In Proceedings of MIE'97 Medical Informatics Europe, 14th International Congress, Porto Carras, Greece, May 25-29, 1997
5. Jelena Bozek, Mario Mustra, Kresimir Delac, and Mislav Grgic. A Survey of Image Processing Algorithms in Digital Mammography.
6. "World Cancer Report". International Agency for Research on Cancer, 2008.
7. Mathworks R2014b Online Documentation Online [http://www. mathworks. in/help/stats/classify. html].
8. G. Castellano, L. Bonilha, L. M. Li, F. Cendes Texture analysis of medical images, Neuroimage Laboratory, Faculty of Medical Sciences, State University of Campinas, Brazil. Clinical Radiology (2004) 59, 1061–1069.
9. J Suckling et al, The Mammographic Image Analysis Society Digital Mammogram Database Exerpta Medica. International Congress Series, 1994 1069 pp 375-378.
10. Michael Brennan, Comparison of automated feature extraction methods for image based screening of cancer cells, Uppsala Universitat, January 2012.
11. Steven A. Hane, High Content Screening: Science, Techniques and Applications, Wiley Publishers,2008, pp. 66
12. Sharma et al. , Mathematical Modeling In Geographical Information System (gis) & Gps An Overview, Concept Publishing Company, 2006, pp 56.
13. Robert M. Haralick, Statistical and Structural Approaches to Texture, Proceedings of the IEEE, Vol. 67, No. 5, May 1979
14. M. Partio et al. , Rock texture retrieval using gray level co-occurrence matrix, 5th Nordic Signal Processing Symp. , 2002
15. V. Bino Sebastian at al. , Gray level co-occurrence matrices: generalisation and some new features, International Journal of Computer Science Engineering and Information Technology, 2 (2012), pp. 151–157
16. Classification toolbox for MATLAB, Milano Chemometrics and QSAR Research Group, found at: http://michem. disat. unimib. it/chm/download/softwares/help_classification/theory. htm

Mammography Texture Normal Cancerous