A Computer Aided Detection (CAD) System for Microcalcifications in Mammograms - MammoScan mCaD

Clusters of microcalcifications in mammograms are an important sign of breast cancer. This paper presents a complete Computer Aided Detection (CAD) scheme for automatic detection of clustered microcalcifications in digital mammograms. The proposed system, MammoScan μCaD, consists of three main steps. Firstly all potential microcalcifications are detected using a a method for feature extraction, VarMet, and adaptive thresholding. This will also give a number of false detections. The goal of the second step, Classifier level 1, is to remove everything but microcalcifications. The last step, Classifier level 2, uses learned dictionaries and sparse representations as a texture classification technique to distinguish single, benign microcalcifications from clustered microcalcifications, in addition to remove some remaining false detections. The system is trained and tested on true digital data from Stavanger University Hospital, and the results are evaluated by radiologists. The overall results are promising, with a sensitivity > 90 % and a low false detection rate (approx 1 unwanted pr. image, or 0.3 false pr. image).

Authors:

• Kjersti Engan
• Thor Ole Gulsrud
• Karl Fredrik Fretheim
• Barbro Furebotten Iversen
• Liv Eriksen

Keywords:

• mammogram
• microcalcifications
• detection
• CAD
• MammoScan μCaD
• VarMet
• dictionary learning
• texture
• FTCM
• classification
• adaptive thresholding

References:

[1] www.kreftregisteret.no.
[2] H. P. Chan, K. Doi, C. Vyborny, R. Scmidt, C. Metz, K. L. Lam,
T. Ogura, Y. Wu, and H. Macmahon, "Improvement in radiologists-
detection of clustered microcalcifications on mammograms. the potential
of computer-aided diagnosis," Investigative Radiology, vol. 25, no. 10,
pp. 1102-1110, 1990.
[3] H. D. Cheng, Y. M. Lui, and R. Freimanis, "A novel approach to
microcalcification detection using fuzzy logic technique," IEEE Trans.
Medical Imaging, vol. 17, no. 3, pp. 442-450, 1998.
[4] W. J. H. Veldkamp and N. Karssemeijer, "An improved method for detection
of microcalcification clusters in digital mammograms," in Proc.
of SPIE International Symposium Medical Imaging, Image Processing
1999, vol. 3661, San Diego, Ca, USA, May 1999, pp. 512-522.
[5] S. Yu and L. Guan, "A cad system for the automatic detection of
clustered microcalcifcations in digitized mammogram films," IEEE
Trans. Medical Imaging, vol. 19, no. 2, pp. 115-126, 2000.
[6] I. El-Naqa, Y. Yang, M. W. N.P., Galatsanos, and R. Nishikawa, "A
support vector machine approach for detection of microcalcification,"
IEEE Trans. Medical Imaging, vol. 21, no. 12, pp. 1552-1563, 2002.
[7] R. Nishikawa, M. Giger, K. Doi, C. Vyborny, and R. A. Schmidt,
"Computer aided detection of clustered microcalcifications in digital
mammograms," Med. Biol. Eng. Compu., vol. 33, pp. 174-178, 1995.
[8] G. Lemaur, K. Drouiche, and J. DeConinck, "Highly regular wavelets
for the detection of clustered microcalcification in mammograms," IEEE
Trans. Medical Imaging, vol. 22, no. 3, pp. 393-401, 2003.
[9] P.Heinlein, J. Drexl, and W. Schneider, "Integrated wavelets for enchancement
of microcalcifications in digital mammography," IEEE
Trans. Medical Imaging, vol. 22, no. 3, pp. 402-413, 2003.
[10] P. Zhang, B. Verma, and K. Kumar, "A neural-genetic algorithm for
feature selection and breast abnormality classification in digital mammography,"
in Proceedings of IEEE International Joint Conference on
Neural Networks, 2004, vol. 3, 2004, pp. 2003-2308.
[11] H. Yoshida, "Matching pursuit with optimally weighted wavelet packets
for extraction of microcalcifications in mammograms," Applied Signal
Processing, vol. 5, no. 3, pp. 127-141, 1998.
[12] G. Horvath, J. Valyon, G. Strausz, M. Pataki, L. Sragner, L. Lasztovicza,
and N. Szekely, "Intelligent advisory system for screening mammography,"
in Proceedings of IEEE Instrumentation and Measurement
Technology Conference, IMTC 04, vol. 3, May 2004, pp. 2071-2076.
[13] D. A. A. Vega-Corona, "Cad system for identification of microcalcifications
in digitized mammography applying grnn neural networks," in
Proceedings of World Automation Congress, 2004, vol. 17, 2004.
[14] K. Engan and T. O. Gulsrud, "VarMet - a method for detection of image
singularities with application to mammography," WSEAS Transactions
on Signal Processing, vol. 2, no. 9, pp. 1222-1229, Sept. 2006.
[15] J.-L. Starck, "Non Linear Multiscale Transforms," in Multiscale and
Multiresolution Methods, T. Barth, T. Chan, and R. Haimes, Eds.
Springer-Verlag, 2002, pp. 239-278.
[16] R. Gonzalez and R. Woods, Digital Image Processing. USA: Addison
Wesley, 1993.
[17] K. Skretting and J. H. Hus├©y, "Texture classification using sparse
frame-based representations," EURASIP Journal on Applied Signal
Processing, vol. 2006, pp. Article ID 52 561, 11 pages, 2006,
doi:10.1155/ASP/2006/52561.
[18] K. Engan, K. Skretting, and J. Hus├©y, "A family of iterative LS-based
dictionary learning algorithms, ILS-DLA, for sparse signal representation,"
Digital Signal Processing, Elsevier, vol. 17, no. 1, pp. 32-49,
2007, doi:10.1016/j.dsp.2006.02.002.
[19] K. Engan, K. Skretting, J. Herredsvela, and T. Gulsrud, "Frame
texture classification method (FTCM) applied on mammograms for
detection of abnormalities," WASET International Journal of Signal
Processing (IJSP), vol. 4, no. 2, 2007, iSSN = 1304-4478,
http://www.waset.org/ijsp/v4/v4-2-16.pdf.
[20] M. Vetterli and J. Kovaˇcevi'c, Wavelets and Subband Coding. Englewood
Cliffs: Prentice-Hall, 1995.
[21] B. K. Natarajan, "Sparse approximate solutions to linear systems," SIAM
journal on computing, vol. 24, pp. 227-234, Apr. 1995.
[22] M. Gharavi-Alkhansari and T. S. Huang, "A fast orthogonal matching
pursuit algorithm," in Int. Conf. on Acoust. Speech and Signal Proc.,
Seattle, U.S.A, May 1998, pp. 1389-1392.
[23] S. F. Cotter, J. Adler, B. D. Rao, and K. Kreutz-Delgado, "Forward
sequential algorithms for best basis selection," IEE Proc. Vis. Image
Signal Process, vol. 146, no. 5, pp. 235-244, Oct. 1999.
[24] K. Skretting, K. Engan, and J. Hus├©y, "Ecg compression using signal
dependent frames and matching pursuit," in Proc. Int. Conf. Acoust.
Speech, Signal Proc., Philadelphia, Pennsylvania, USA, 2005.
[25] A.Rahmoune, P. Vandergheynst, and P. Frossard, "MP3D: Highly sclable
video coding scheme based on matching pursuit," in Proc. Int. Conf.
Acoust. Speech, Signal Proc., Montreal, Canada, May 2004.
[26] K. Engan, K. Skretting, and J. Hus├©y, "Denoising of images using signal
dependent frames and matching pursuit," in Proc. Int. Conf. Acoust.
Speech, Signal Proc., Philadelphia, Pennsylvania, USA, 2005.
[27] T. W. Lee, M. S. Lewicki, M. Girolami, and T. J. Sejnowski, "Blind
source separation of more sources than mixtures using overcomplete
representaions," IEEE Signal Processing Letters, vol. 6, no. 4, pp. 87-
90, Apr. 1999.
[28] P. Bofill and M. Zibulevsky, "Underdetermined blind source separation
using sparse represenations," Signal Processing, vol. 81, no. 11, pp.
2353-2362, 2001.
[29] M.Zibulevsky and B. Pearlmutter, "Blind source separation by sparse
decomposition in a signal dictionary," Neural Computation, no. 13, pp.
863-882, 2001.
[30] K. Engan, S. O. Aase, and J. H. Hus├©y, "Method of optimal directions
for frame design," in Proc. ICASSP -99, Phoenix, USA, Mar. 1999, pp.
2443-2446.
[31] ÔÇöÔÇö, "Multi-frame compression: Theory and design," Signal Processing,
vol. 80, pp. 2121-2140, Oct. 2000.
[32] M. Tuceryan and A. K. Jain, "Texture analysis," in Handbook of Pattern
Recognition and Computer Vision, C. H. Chen, L. F. Pau, and P. S. P.
Wang, Eds. Singapore: World Scientific Publishing Co, 1998, ch. 2.1,
pp. 207-248.
[33] www.r2tech.com.