Abstract
Mammograms are X-ray images of human breast which are normally used to detect breast cancer. The presence of pectoral muscle in mammograms may disturb the detection of breast cancer as the pectoral muscle and mammographic parenchyma appear similar. So, the suppression or exclusion of the pectoral muscle from the mammograms is demanded for computer-aided analysis which requires the identification of the pectoral muscle. The main objective of this study is to propose an automated method to efficiently identify the pectoral muscle in medio-lateral oblique-view mammograms. This method uses a proposed graph cut-based image segmentation technique for identifying the pectoral muscle edge. The identified pectoral muscle edge is found to be ragged. Hence, the pectoral muscle is smoothly represented using Bezier curve which uses the control points obtained from the pectoral muscle edge. The proposed work was tested on a public dataset of medio-lateral oblique-view mammograms obtained from mammographic image analysis society database, and its performance was compared with the state-of-the-art methods reported in the literature. The mean false positive and false negative rates of the proposed method over randomly chosen 84 mammograms were calculated, respectively, as 0.64% and 5.58%. Also, with respect to the number of results with small error, the proposed method out performs existing methods. These results indicate that the proposed method can be used to accurately identify the pectoral muscle on medio-lateral oblique view mammograms.
Similar content being viewed by others
References
American cancer society: Cancer facts and figures 2005, Atlanta: American Cancer Society, 2005
Eklund GW, Cardenosa G, Parsons W: Assessing adequacy of mammographic image quality. Radiology 190(2):297–307, 1994
Zhou C, Hadjiiski LM, Paramagul C, Sahiner B, Chan H-P, Wei J: Computerized pectoral muscle identification on MLO-view mammograms for CAD applications. Proc of the SPIE 5747:852–857, 2005
Kwok SM, Chandrasekhar R, Attikiouzel Y, Rickard MT: Automatic pectoral muscle segmentation on mediolateral oblique view mammograms. IEEE Trans Med Imaging 23(9):1129–1140, 2004
Ferrari RJ, Rangayyan RM, Desautels JEL, Borges RA, Frère AF: Automatic identification of the pectoral muscle in mammograms. IEEE Trans Med Imaging 23(2):232–245, 2004
Suckling J, Dance DR, Moskovic E, Lewis DJ, Blacker SG: Segmentation of mammograms using multiple linked self-organizing neural networks. Med Phys 22(2):145–152, 1995
Masek M, Chandrasekhar R, Desilva CJS, and Attikiouzel Y: Spatially based application of the minimum cross-entropy thresholding algorithm to segment the pectoral muscle in mammograms, The Seventh Australian and New Zealand Intelligent Information Systems Conference, Nov. 18–21. 101–106, 2001
Brink AD, Pendock NE: Minimum cross entropy threshold selection. Pattern Recogn 27(1):179–188, 1996
Kwok SM, Chandrasekhar R, and Attikiouzel Y: Automatic pectoral muscle segmentation on mammograms by straight line estimation and cliff detection. The Seventh Australian New Zealand Intelligent information Systems Conference, Perth, Western Australia, Nov. 18–21. 2001
Karssemeijer N: Automated classification of parenchymal patterns in mammograms. Phys Med Biol 43(2):365–378, 1998
Ferrari RJ, Rangayyan RM, Desautels JEL, and Frere AF: Segmentation of mammograms: Identification of the skin-air boundary, pectoral muscle, and fibro-glandular disc. In IWDM 2000: Proc. 5th Int. Workshop Digital Mammography. 573–579, June 2001
Yam M, Brady M, Highnam R, Behrenbruch C, English R, Kita Y: Three-dimensional reconstruction of microcalcification clusters from two mammographic views. IEEE Trans Med Imaging 20:479–489, 2001
Aylward SR, Hemminger BM, Pisano ED: Mixture modeling for digital mammogram display and analysis. In Digital Mammography. Comput Imaging Vis 13:305–312, 1998
Weidong X, and Shunren X: A model based algorithm to segment the pectoral muscle in mammograms. IEEE Int. Conf. Neural Networks & Signal Processing, Nanjing, China, Dec.14–17. 1163–1169, 2003
Raba D, Oliver A, Martí J, Peracaula M, Espunya J: Breast segmentation with pectoral muscle suppression on digital mammograms. Lect Notes Comp Sci 3523:471–478, 2005
Bajger M, Ma F, Bottema MJ: Minimum spanning trees and active contours for identification of the pectoral muscle in screening mammograms. Digital Image Computing: Techniques and Applications conference. 47–53, 2005
Ma F, Bajger M, Bottema MJ: Extracting the pectoral muscle in screening mammograms using a graph pyramid. APRS workshop on digital image computing, University of Queensland. 39–42, 2005
Ma F, Bajger M, Slavotinek JP, Bottema MJ: Two graph theory based methods for identifying the pectoral muscle in mammograms. Pattern Recogn 40:2592–2602, 2007
Bassett LW, Hirbawi IA, Debruhl N, Hayes MK: Mammographic positioning: Evaluation from the view box. Radiology 188:803–806, 1993
Chandrasekhar R, Kwok SM, and Attikiouzel Y: Automatic evaluation of mammographic adequacy and quality on the mediolateral oblique view. In Digital Mammography: IWDM 2002: Proc. 6th Int. Workshop Digital Mammography, Bremen, Germany, June 22–25. 182–186, 2002
Tucker AB: Computer science handbook, 2nd edition. Boca Raton: CRC, 2004
Suckling J, Parker J, Dance DR, Astley S, Hutt I, Boggis CRM, Ricketts I, Stamatakis E, Cerneaz N, Kok SL, Taylor P, Betal D, Savage J: The Mammographic Image Analysis Society Digital Mammogram Database. Excerpta Medica International Congress Series 1069:375–378, 1994
Cormen TH, Leiserson CE, Rivest RL, Stein C: Introduction to algorithms, 2nd edition. Boston: MIT, 2001
Loudon K: Mastering algorithms with C, Sebastopol: O'Reilly, 1999
Acknowledgement
The authors wish to thank the anonymous reviewers for their important corrections and suggestions that have been included in the text. The authors would like to thank Rangayyan RM for providing a set of mammograms with radiologist drawn pectoral muscle boundaries. The authors would also like to thank Vimal SP, of BITS Pilani, for the useful discussions with him.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
The merge criterion can be expanded as two conditions as
Let \(K = 2 \times N_{R} \). Now, Eqs. 17a and 17b can be written as
For the merge criterion to be true, any one of these conditions in Eqs. 18a and 18b must be satisfied. If the two regions, R 1 and R 2, are homogeneous, then the non-homogeneity measures \({\text{IRM}}{\left( {R_{1} ,\,R_{2} } \right)} - {\text{IRA}}{\left( {R_{i} } \right)},\,i \in {\left\{ {1,2} \right\}}\) become small; hence, any one of these conditions is more likely to be satisfied. However, if the regions are not homogeneous, then the non-homogeneity measure values become high and also, the values are further enhanced by the multiplicative factor, the size of the region; hence, both these conditions are not satisfied.
Rights and permissions
About this article
Cite this article
Camilus, K.S., Govindan, V.K. & Sathidevi, P.S. Computer-Aided Identification of the Pectoral Muscle in Digitized Mammograms. J Digit Imaging 23, 562–580 (2010). https://doi.org/10.1007/s10278-009-9240-6
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10278-009-9240-6