Massive Lesions Classification using Features based on Morphological Lesion Differences

Authors: U. Bottigli, D.Cascio, F. Fauci, B. Golosio, R. Magro, G.L. Masala, P. Oliva, G. Raso, S.Stumbo

Abstract:

Purpose of this work is the development of an automatic classification system which could be useful for radiologists in the investigation of breast cancer. The software has been designed in the framework of the MAGIC-5 collaboration. In the automatic classification system the suspicious regions with high probability to include a lesion are extracted from the image as regions of interest (ROIs). Each ROI is characterized by some features based on morphological lesion differences. Some classifiers as a Feed Forward Neural Network, a K-Nearest Neighbours and a Support Vector Machine are used to distinguish the pathological records from the healthy ones. The results obtained in terms of sensitivity (percentage of pathological ROIs correctly classified) and specificity (percentage of non-pathological ROIs correctly classified) will be presented through the Receive Operating Characteristic curve (ROC). In particular the best performances are 88% ± 1 of area under ROC curve obtained with the Feed Forward Neural Network.

Keywords: Neural Networks, K-Nearest Neighbours, SupportVector Machine, Computer Aided Diagnosis.

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1083575

References:

[1] Smith R.A., "Epidemiology of breast cancer", in "A categorical course in physics. Imaging considerations and medical physics responsibilities", Madison, Winsconsin, Medical Physics Publishing, 1991.
[2] Peto R., Boreham J., Clarke m., Davies c., Deral V, correspondence "UK and USA Breast cancer deaths down 25% in year 2000 at ages 20-69 years", LANCET 2000, 355, (9217) pp. 1822-1823, 2000.
[3] Bird R., Wallace T., Yankaskas B., "Analysis of cancer missed at screening mammography", Radiology 1992: 184; pp.613-617, 1992.
[4] Bottigli U, Delogu P, Fantacci ME, Fauci F, Golosio B, Lauria A, Palmiero R, Raso G, Stumbo S, Tangaro S Search of Microcalcification clusters with the CALMA CAD station. The International Society for Optical Engineering (SPIE) 4684: 1301-1310, 2002
[5] F. Fauci, S. Bagnasco, R. Bellotti, D. Cascio, S. C. Cheran, F. De Carlo, G. De Nunzio, M. E. Fantacci, G. Forni, A. Lauria, E.Lopez Torres, R. Magro, G. L. Masala, P.Oliva, M. Quarta, G. Raso, A. Retico, S.Tangaro, Mammogram Segmentation by Contour Searching and Massive Lesion Classification with Neural Network, Proc. IEEE Medical Imaging Conference, October 16-22 2004, Rome, Italy; M2- 373/1-5, 2004.
[6] O. Duda, P. E. Hart, D. G. Stark, "Pattern Classification", second edition, A Wiley-Interscience Publication John Wiley & Sons, 2001.
[7] S. J. Russel, P.Norvig, "Artificial Intelligence. A modern approach", UTET, 1998.
[8] S. Haykin "Neural Networks - A comprehensive foundation", second edition, Prentice Hall, 1999.
[9] V. N. Vapnik. "Statistical Learning Theory. Wiley", New York , 1998.
[10] M. Pontil, A. Verri "Properties of Support Vector Machines", Neural Computation, Vol. 10, pp 955-974, 1998.
[11] N. Cristianini, J. Shave-Taylor. "An Introduction to Support Vector Machine"(and other kernel-based learning methods). Cambridge University Press 2000.
[12] SVM_light software is available in the following location : ftp://ftp-ai.cs.unidortmund.de/pub/Users/thorsten/svm_light/current/ svm_light.tar.gz
[13] T. Joachims, Text Categorization with Support Vector Machines: Learning with Many Relevant Features, Proc. 10th European Conf. Machine Learning (ECML), Springer-Verlag, 1998.
[14] T. Mitchell "Machine Learning" , McGraw-Hill 1997.
[15] Timp S., Karssemeijer N., A new 2D segmentation method based on dynamic programming applied to computer aided detection in mammography, Medical Physics: 31; 958-971, 2004.
[16] Baydush A.H., Catarious D.M., Abbey C.K., Floyd C.E., Computer aided detection of masses in mammography using subregion Hotelling observers, Medical Physics: 30; 1781-1787, 2003.
[17] Tourassi G.D., Vargas-Voracek R., Catarious D.M. Jr, Floyd C.E. Jr, Computer-assisted detection of mammographic masses: A template matching scheme based on mutual information, Medical Physics: 30 (8); 2123-2130, 2003.
[18] Antonie M.L., Zaiane O.R., Coman A., Application of data mining techniques for medical image classification, Proc. of II Int. Work. On Multimedia Data Mining, USA, 2001.
[19] Vyborny CJ., Giger ML., Computer vision and artificial intelligence in mammography, AJR: 162; 699-708, 1994.
[20] Lai S., Li X., Bischof W., On techniques for detecting circumscribed masses in mammograms", IEEE Transaction on Medical Imaging: 8(4); 377-386, 1989.
[21] Hanley JA, McNeil B, The meaning and use of the area under a receiver operating characteristic (ROC) curve, Radiology: 143; 29-36, 1982.
[22] Hanley JA, McNeil B, A method of comparing the areas under receiver operating characteristic curves derived from the same cases, Radiology: 148; 839-843, 1983.
[23] U. Bottigli, B. Golosio, G. L. Masala, P. Oliva, S. Stumbo, D. Cascio, F. Fauci, R. Magro, G. Raso, R. Bellotti, F. De Carlo, S.Tangaro, I. De Mitri, G. De Nunzio, M. Quarta, A. Preite Martinez, P. Cerello, S. C. Cheran, E.Lopez Torres "Dissimilarity Application for Medical Imaging Classification" on proceedings of The 9th World Multi-Conference on Systemics, Cybernetics and Informatics WMSCI 2005, Orlando 10-13 July 2005, vol III pag 258-262, 2005.
[24] G. Masala, B. Golosio, D. Cascio, F. Fauci, S. Tangaro, M. Quarta, S. C Cheran, E. L. Torres, "Classifiers trained on dissimilarity representation of medical pattern: a comparative study" on Nuovo Cimento C, Vol 028, Issue 06, pp 905-912 , 2005.