Abstract
This paper proposes a method to segment the pectoralis muscles in CT scans of patients within a Bayesian framework. First, a subject-tailored probabilistic atlas is constructed using affine registered label-maps of subjects that are highly similar to the test subject from a database of pairwise registered training subjects. The likelihood is constructed using a multivariate distribution taking intensities and distance to the atlas into account. The posterior probability is used to drive a graph cuts segmentation for classifying the CT into left major, left minor, right major, right minor pectoralis and non-pectoralis taking neighborhood information into account. The probabilistic prior is built using 400 CT scans and the method is tested on 50 independent CT scans. Results are reported on each muscle separately and show a statistically significant improvement when the subject-tailored prior is incorporated into the model. This automatic method can be used to objectively and efficiently asses clinical outcomes for patients with COPD.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Schols, A.M., et al.: Body composition and mortality in chronic obstructive pulmonary disease. Am. J. Clin. Nutr. 82(1), 53–59 (2005)
Kim, S., et al.: Body mass index and fat free mass index in obstructive lung disease in korea. Int. J. Tuberc. Lung Dis. 18(1), 102–108 (2014)
McDonald, M.L.N., et al.: Quantitative computed tomographic measures of pectoralis muscle area and disease severity in chronic obstructive pulmonary disease: A cross-sectional study. Ann. Am. Thorac. Soc. 11, 326–334 (2014)
Ganesan, K., et al.: Pectoral muscle segmentation: a review. Comput. Meth. Programs Biomed. 110(1), 48–57 (2013)
Shimizu, A., Nakagomi, K., Narihira, T., Kobatake, H., Nawano, S., Shinozaki, K., Ishizu, K., Togashi, K.: Automated segmentation of 3D CT images based on statistical atlas and graph cuts. In: Menze, B., Langs, G., Tu, Z., Criminisi, A. (eds.) MICCAI 2010. LNCS, vol. 6533, pp. 214–223. Springer, Heidelberg (2011)
Park, H., Bland, P.H., Hero III, A.O., Meyer, C.R.: Least biased target selection in probabilistic atlas construction. In: Duncan, J.S., Gerig, G. (eds.) MICCAI 2005. LNCS, vol. 3750, pp. 419–426. Springer, Heidelberg (2005)
Blezek, D.J., Miller, J.V.: Atlas stratification. MIA 11(5), 443–457 (2007)
Gubern-Mérida, A., Kallenberg, M., Martí, R., Karssemeijer, N.: Segmentation of the pectoral muscle in breast mri using atlas-based approaches. In: Ayache, N., Delingette, H., Golland, P., Mori, K. (eds.) MICCAI 2012, Part II. LNCS, vol. 7511, pp. 371–378. Springer, Heidelberg (2012)
Boykov, Y., et al.: Fast approximate energy minimization via graph cuts. IEEE Trans. Pattern Anal. Mach. Intell. 23(11), 1222–1239 (2001)
Acknowledgements
This work has been supported by NIH NHLBI award number 1R01HL116931, 1R01HL122464-01A1 and K25-HL104085. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Harmouche, R., Ross, J.C., Washko, G.R., San José Estépar, R. (2014). Pectoralis Muscle Segmentation on CT Images Based on Bayesian Graph Cuts with a Subject-Tailored Atlas. In: Menze, B., et al. Medical Computer Vision: Algorithms for Big Data. MCV 2014. Lecture Notes in Computer Science(), vol 8848. Springer, Cham. https://doi.org/10.1007/978-3-319-13972-2_4
Download citation
DOI: https://doi.org/10.1007/978-3-319-13972-2_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-13971-5
Online ISBN: 978-3-319-13972-2
eBook Packages: Computer ScienceComputer Science (R0)