Abstract
Automatic 3D liver segmentation is a fundamental step in the liver disease diagnosis and surgery planning. This paper presents a novel fully automatic algorithm for 3D liver segmentation in clinical 3D computed tomography (CT) images. Based on image features, we propose a new Mahalanobis distance cost function using an active shape model (ASM). We call our method MD-ASM. Unlike the standard active shape model (ST-ASM), the proposed method introduces a new feature-constrained Mahalanobis distance cost function to measure the distance between the generated shape during the iterative step and the mean shape model. The proposed Mahalanobis distance function is learned from a public database of liver segmentation challenge (MICCAI-SLiver07). As a refinement step, we propose the use of a 3D graph-cut segmentation. Foreground and background labels are automatically selected using texture features of the learned Mahalanobis distance. Quantitatively, the proposed method is evaluated using two clinical 3D CT scan databases (MICCAI-SLiver07 and MIDAS). The evaluation of the MICCAI-SLiver07 database is obtained by the challenge organizers using five different metric scores. The experimental results demonstrate the availability of the proposed method by achieving an accurate liver segmentation compared to the state-of-the-art methods.
Acknowledgments
The work was partially funded by DAAD scholarship (A/10/96106) and MOHESR-Iraq (Baghdad University). The authors gratefully acknowledge these supports.
References
[1] Aharon M, Elad M, Bruckstein A. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation. IEEE Trans Signal Process 2006; 54: 4311–4322.10.1109/TSP.2006.881199Search in Google Scholar
[2] Badakhshannoory H, Saeedi P. Automatic liver segmentation from CT scans using multi-layer segmentation and principal component analysis. In: MICCAI proceeding. LNCS; vol. 6454, 2010: 342–350.10.1007/978-3-642-17274-8_34Search in Google Scholar
[3] Baldock R, Cootes TF. Model-based methods in analysis of biomedical images. In: Baldock R, Graham J, editors. Image processing and analysis. UK: Oxford University Press, Chapter 7, 1999: 223–248.Search in Google Scholar
[4] Boltcheva D, Yvinec M, Boissonnat JD. Mesh generation from 3D multi-material images. In: MICCAI. LNCS; vol. 5762, 2009: 283–290.10.1007/978-3-642-04271-3_35Search in Google Scholar PubMed
[5] Boykov YY, Jolly MP. Interactive graph cuts for optimal boundary and region segmentation of objects in N-D images. In: ICCV. Vancouver, Canada: IEEE 2001: 105–112.Search in Google Scholar
[6] Campadelli P, Casiraghi E, Esposito A. Liver segmentation from computed tomography scans: a survey and a new algorithm. Artif Intell Med 2009; 45: 185–196.10.1016/j.artmed.2008.07.020Search in Google Scholar PubMed
[7] Chen Y, Zhao W, Wu Q, Wang Z, Hu J. Liver segmentation in CT images for intervention using a graph-cut based model. MICCAI proceeding. LNCS; vol. 7029, 2012: 157–164.10.1007/978-3-642-28557-8_20Search in Google Scholar
[8] Chi Y, Cashman PMM, Bello F, Kitney RI. A discussion on the evaluation of a new automatic liver volume segmentation method for specified CT image datasets. In: MICCAI-3D segmenttion in the clinic: a grand challenge. Brisbane, Australia: MICCAI Workshop SLiver07 2007: 167–175.Search in Google Scholar
[9] Gauriau R, Cuingnet R, Prevost R, et al. A generic, robust and fully-automatic workflow for 3D CT liver segmentation. In: MICCAI Abdominal Imaging. LNCS; vol. 8198, 2013: 241–250.10.1007/978-3-642-41083-3_27Search in Google Scholar
[10] Haralick RM, Shanmugam K, Dinstein I. Texture feature for image classification. IEEE Trans Syst Man Cybernet 1973; SMC-3(6): 610–621.10.1109/TSMC.1973.4309314Search in Google Scholar
[11] Heimann T, Styner M, van Ginneken B. Sliver07. http://www.sliver07.org (2007).Search in Google Scholar
[12] Heimann T, Thorn M, Kunert T, Meinzer HP. New methods for leak detection and contour correction in seeded region growing segmentation. In: ISPRS Congress proceeding. Istanbul: The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences 2004: 317–322.Search in Google Scholar
[13] Heimann T, van Ginneken B, Styner M, et al. Comparison and evaluation of methods for liver segmentation from CT datasets. IEEE Trans Med Imag 2009; 28: 1251–1265.10.1109/TMI.2009.2013851Search in Google Scholar PubMed
[14] Huang C, Jia F, Fang C, Hu Q. Automatic liver detection and segmentation from 3D CT images: a hybrid method using statistical pose model and probabilistic atlas. Int J Comput Assist Radiol Surg 2013; 8: 237–238.Search in Google Scholar
[15] Huang C, Jia F, Li Y, et al. Automatic liver segmentation based on shape constrained differeomorphic demons atlas registration. In: ICECC proceedings of the international conference on electronics, communications and control; 2012: 126–129.Search in Google Scholar
[16] Huang C, Li X, Jia F. Automatic liver segmentation using multiple prior knowledge models and free-form deformation. In: Proceedings of the VISCERAL organ segmentation and landmark detection benchmark at the 2014 IEEE international symposium on biomedical imaging (ISBI); 2014. http://ceur-ws.org.Search in Google Scholar
[17] Kainmueller D, Lange T, Lamecker H. Shape constrained automatic segmentation of the liver based on a heuristic intensity model. In: MICCAI-3D segmenttion in the clinic: a grand challenge. Brisbane, Australia: MICCAI Workshop SLiver07 2007: 109–116.Search in Google Scholar
[18] Kirschner M, Oost E, Shimizu A, Watanabe H, Nawano S. An application of the probabilistic active shape model to liver segmentation in CT images. MICCAI-SLiver, 2013. http://sliver07.org/data/2013-04-11-1729.pdf.Search in Google Scholar
[19] Linguraru MG, Richbourg WJ, Watt JM, Pamulapati V, Summers RM. Liver and tumor segmentation and analysis from CT of diseased patients via a generic affine invariant shape parameterization and graph cuts. In: MICCAI abdominal imaging. LNCS; vol. 7029, 2011: 198–206.10.1007/978-3-642-28557-8_25Search in Google Scholar
[20] MIDAS/NLM: Imaging methods assessment and reporting/public data standards. http://www.insight-journal.org/midas/collection/view/38.Search in Google Scholar
[21] Tomoshige S, Oost E, Shimizu A, Watanabe H, Nawano S. A Conditional statistical shape model with integrated error estimation of the conditions; application to liver segmentation in non-contrast CT images. Med Image Anal 2014; 18: 130–143.10.1016/j.media.2013.10.003Search in Google Scholar PubMed
[22] Tustison N, Gee J. Introducing dice, jaccard, and other label overlap measures to ITK. Insight J 2009; 1–4. Available at: http://hdl.handle.net/10380/3141.10.54294/1vixggSearch in Google Scholar
[23] Wang W, Shan S, Gao W, Cao B, Yin B. An improved active shape model for face alignment. In: Proceedings of IEEE international conference on multimodal interfaces; 2002: 523–528.Search in Google Scholar
[24] Wang F, Vemuri BC, Rangarajan A, Eisenschenk SJ. Simultaneous nonrigid registration of multiple point sets and atlas construction. IEEE Trans Pattern Anal Mach Intell 2008; 30: 2011–2022.10.1007/11744078_43Search in Google Scholar
[25] Wimmer A, Hornegger J, Soza G. Implicit active shape model employing boundary classifier. In: IEEE ICPR. Florida, USA: IEEE 2008: 1–4.10.1109/ICPR.2008.4760968Search in Google Scholar
[26] Wimmer A, Soza G, Hornegger J. A Generic probabilistic active shape model for organ segmentation. In: MICCAI proceeding. LNCS; vol. 5762, 2009: 26–33.10.1007/978-3-642-04271-3_4Search in Google Scholar PubMed
[27] Xing EP, Ng AY, Jordan MI, Russell S. Distance metric learning, with application to clustering with side-information. Cambridge: Advances in Neural Information Processing Systems, MIT Press, Vol. 15, 2003: 505–512.Search in Google Scholar
©2016 Walter de Gruyter GmbH, Berlin/Boston
