In this thesis, we have proposed a computational framework that automatically segments the left ventricle in short-axis (SAX) view cine cardiac magnetic resonance (CMR) images. We employ novel cost-volume filtering (CVF) scheme combined with novel myocardial contour processing framework to overcome the segmentation difficulty resulted from PMTC (papillary muscle and trabeculae carneae) tissues that prevents previous methods from achieving high segmentation accuracy. Experimental result shows improved accuracy and robustness over previous works. Highlights regarding this work are:
• The first work to use CVF in the context of medical image segmentation.
Furthermore, we have proposed a novel cost initialization scheme specifically tailored to CMR images for improved accuracy.
• The contour processing framework processes the 1D contour function transformed from the segmentation result produced by CVF. By exploiting gradient information, a complimentary contour is also generated. In the final stage, two raw contours obtained from region-based CVF and gradient-based Canny’s edge detector undergo a combination and regularization process in which suitable weighting function for each raw contour are determined and additional constraints are enforced.
• One of the issues in previous left ventricle segmentation methods is erroneously including the papillary muscle and trabeculae as a part of the myocardium. A new constraint in contour regularization process is proposed to ensure exclusion of them. To our knowledge, no other works have used similar methods nor do they consider this piece of information to improve myocardial contour accuracy.
• Our method is robust against various pathologies. Specifically, a total of 45 subjects spanning three pathological plus one healthy cases are tested. All four cases show improvement in segmentation accuracy and in derived cardiac parameters.
Furthermore, given that many information are yet to be exploited, such as the inherent 3D information in CMR volume and heart movement constraints, we believe the performance of the proposed segmentation algorithm can be improved much further once all the information is taken into consideration. In addition to the quantitative analysis that shows close correlation between manual and automated segmentation, the high segmentation accuracy and low manual intervention rate suggest our work has potential to accurate 3D reconstruction of left ventricle for visualizing the shape and motion of the left ventricle. Future work includes exploiting inter-slice and inter-time relationships in order to optimize the contour globally, as well as extending this methodology to the delineation of left ventricular epicardial contour, which will make additional clinical parameters such as myocardial mass available.
REFERENCE
[1] Global status report on noncommunicable diseases 2010. Geneva, World Health Organization, 2011.
[2] P. A. Heidenreich, J. G. Trogdon, O. A. Khavjou, J. Butler, K. Dracup, M. D.
Ezekowitz, E. A. Finkelstein, Y. Hong, S. C. Johnston, A. Khera, D. M. Lloyd-Jones, S.
A. Nelson, G. Nichol, D. Orenstein, P. W. F. Wilson, and Y. J. Woo, “Forecasting the Future of Cardiovascular Disease in the United States A Policy Statement From the American Heart Association,” Circulation, vol. 123, no. 8, pp. 933–944, Mar. 2011.
[3] K. S. Reddy and S. Yusuf, “Emerging Epidemic of Cardiovascular Disease in Developing Countries,” Circulation, vol. 97, no. 6, pp. 596–601, Feb. 1998.
[4] H.-Y. Lee, N. C. F. Codella, M. D. Cham, J. W. Weinsaft, and Y. Wang, “Automatic Left Ventricle Segmentation Using Iterative Thresholding and an Active Contour Model With Adaptation on Short-Axis Cardiac MRI,” IEEE Trans. Biomed. Eng., vol. 57, no. 4, pp. 905–913, Apr. 2010.
[5] R. J. van der Geest, E. Jansen, V. G. M. Buller, and J. H. C. Reiber, “Automated detection of left ventricular epi- and endocardial contours in short-axis MR images,” in Computers in Cardiology 1994, 1994, pp. 33–36.
[6] Y.-L. Lu, K. A. Connelly, A. J. Dick, G. A. Wright, and P. E. Radau, “Automatic functional analysis of left ventricle in cardiac cine MRI,” Quant. Imaging Med. Surg., vol.
3, no. 4, pp. 200–209, Aug. 2013.
[7] C. Petitjean and J.-N. Dacher, “A review of segmentation methods in short axis cardiac MR images,” Med. Image Anal., vol. 15, no. 2, pp. 169–184, Apr. 2011.
[8] C. A. Cocosco, W. J. Niessen, T. Netsch, E. P. A. Vonken, G. Lund, A. Stork, and
M. A. Viergever, “Automatic image-driven segmentation of the ventricles in cardiac cine MRI,” J. Magn. Reson. Imaging, vol. 28, no. 2, pp. 366–374, Aug. 2008.
[9] S. Xu, C. Pei, and H. Hu, “Endocardium and Epicardium Segmentation in MR Images Based on Developed Otsu and Dynamic Programming,” Sens. Transducers, Mar.
2014.
[10] S. Huang, J. Liu, L. C. Lee, S. K. Venkatesh, L. L. S. Teo, C. Au, and W. L.
Nowinski, “An Image-Based Comprehensive Approach for Automatic Segmentation of Left Ventricle from Cardiac Short Axis Cine MR Images,” J. Digit. Imaging, vol. 24, no.
4, pp. 598–608, Aug. 2011.
[11] Y. Lu, P. Radau, K. Connelly, A. Dick, and G. Wright, “Automatic Image-Driven Segmentation of Left Ventricle in Cardiac Cine MRI,” MIDAS J. - Card. MR Left Ventricle Segmentation Chall., 2009.
[12] S. Huang, J. Liu, L. C. Lee, S. K. Venkatesh, L. L. S. Teo, C. Au, and W. L.
Nowinski, “Segmentation of the Left Ventricle from Cine MR Images Using a Comprehensive Approach,” MIDAS J. - Card. MR Left Ventricle Segmentation Chall., 2009.
[13] J. Cousty, L. Najman, M. Couprie, S. Clément-Guinaudeau, T. Goissen, and J. Garot,
“Segmentation of 4D cardiac MRI: Automated method based on spatio-temporal watershed cuts,” Image Vis. Comput., vol. 28, no. 8, pp. 1229–1243, Aug. 2010.
[14] H. Hu, H. Liu, Z. Gao, and L. Huang, “Hybrid segmentation of left ventricle in cardiac MRI using Gaussian-mixture model and region restricted dynamic programming,”
Magn. Reson. Imaging, vol. 31, no. 4, pp. 575–584, May 2013.
[15] Nobuyuki Otsu, “A Threshold Selection Method from Gray-Level Histograms,”
IEEE Trans. Syst. Man Cybern., vol. 9, no. 1, pp. 62–66, Jan. 1979.
[16] C. Constantinides, E. Roullot, M. Lefort, and F. Frouin, “Fully automated
segmentation of the left ventricle applied to cine MR images: Description and results on a database of 45 Subjects,” in 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2012, pp. 3207–3210.
[17] J. Wijnhout, D. Hendriksen, H. Van Assen, and R. Van der Geest, “LV Challenge LKEB Contribution: Fully Automated Myocardial Contour Detection,” MIDAS J. - Card.
MR Left Ventricle Segmentation Chall., 2009.
[18] I. Ben Ayed, S. Li, and I. Ross, “Embedding Overlap Priors in Variational Left Ventricle Tracking,” IEEE Trans. Med. Imaging, vol. 28, no. 12, pp. 1902–1913, Dec.
2009.
[19] G. Tarroni, D. Marsili, F. Veronesi, C. Corsi, C. Lamberti, and G. Sanguinetti,
“Near-automated 3D segmentation of left and right ventricles on magnetic resonance images,” in 2013 8th International Symposium on Image and Signal Processing and Analysis (ISPA), 2013, pp. 522–527.
[20] M. Lynch, O. Ghita, and P. F. Whelan, “Segmentation of the Left Ventricle of the Heart in 3-D+t MRI Data Using an Optimized Nonrigid Temporal Model,” IEEE Trans.
Med. Imaging, vol. 27, no. 2, pp. 195–203, Feb. 2008.
[21] Q. C. Pham, F. Vincent, P. Clarysse, P. Croisille, and I. E. Magnin, “A FEM-based deformable model for the 3D segmentation and tracking of the heart in cardiac MRI,” in Proceedings of the 2nd International Symposium on Image and Signal Processing and Analysis, 2001. ISPA 2001, 2001, pp. 250–254.
[22] J. Schaerer, C. Casta, J. Pousin, and P. Clarysse, “A dynamic elastic model for segmentation and tracking of the heart in MR image sequences,” Med. Image Anal., vol.
14, no. 6, pp. 738–749, Dec. 2010.
[23] C. Casta, P. Clarysse, J. Schaerer, and J. Pousin, “Evaluation of the Dynamic Deformable Elastic Template model for the segmentation of the heart in MRI sequences,”
MIDAS J. - Card. MR Left Ventricle Segmentation Chall., 2009.
[24] M.-P. Jolly, C. Guetter, X. Lu, H. Xue, and J. Guehring, “Automatic Segmentation of the Myocardium in Cine MR Images Using Deformable Registration,” in Statistical Atlases and Computational Models of the Heart. Imaging and Modelling Challenges, O.
Camara, E. Konukoglu, M. Pop, K. Rhode, M. Sermesant, and A. Young, Eds. Springer Berlin Heidelberg, 2012, pp. 98–108.
[25] M. Lorenzo-Valdés, G. I. Sanchez-Ortiz, A. G. Elkington, R. H. Mohiaddin, and D.
Rueckert, “Segmentation of 4D cardiac MR images using a probabilistic atlas and the EM algorithm,” Med. Image Anal., vol. 8, no. 3, pp. 255–265, Sep. 2004.
[26] V. Hartwig, G. Giovannetti, N. Vanello, M. Lombardi, L. Landini, and S. Simi,
“Biological effects and safety in magnetic resonance imaging: a review,” Int. J. Environ.
Res. Public. Health, vol. 6, no. 6, pp. 1778–1798, Jun. 2009.
[27] A. O. Zurick, J. L. Klein, and M. S. Runge, Netter’s Cardiology, 2nd ed. Elsevier, 2010.
[28] M.-P. Jolly, “Automatic Recovery of the Left Ventricular Blood Pool in Cardiac Cine MR Images,” in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2008, D. Metaxas, L. Axel, G. Fichtinger, and G. Székely, Eds. Springer Berlin Heidelberg, 2008, pp. 110–118.
[29] C. Rhemann, A. Hosni, M. Bleyer, C. Rother, and M. Gelautz, “Fast cost-volume filtering for visual correspondence and beyond,” in 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2011, pp. 3017–3024.
[30] A. Pednekar, U. Kurkure, R. Muthupillai, S. Flamm, and I. Kakadiaris, “Automated left ventricular segmentation in cardiac MRI,” IEEE Trans. Biomed. Eng., vol. 53, no. 7, pp. 1425–1428, Jul. 2006.
[31] K. He, J. Sun, and X. Tang, “Guided Image Filtering,” IEEE Trans. Pattern Anal.
Mach. Intell., vol. 35, no. 6, pp. 1397–1409, Jun. 2013.
[32] J. Canny, “A Computational Approach to Edge Detection,” IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-8, no. 6, pp. 679–698, Nov. 1986.
[33] P. Radau, Y. Lu, K. Connelly, G. Paul, A. J. Dick, and G. A. Wright, “Evaluation Framework for Algorithms Segmenting Short Axis Cardiac MRI.,” MIDAS J. - Card. MR Left Ventricle Segmentation Chall., 2009.
[34] M. Jolly, “Fully automatic left ventricle segmentation in cardiac cine MR images using registration and minimum surfaces,” MIDAS J.-Card. MR Left Ventricle Segmentation Chall., vol. 4, 2009.
[35] C. Feng, C. Li, D. Zhao, C. Davatzikos, and H. Litt, “Segmentation of the Left Ventricle Using Distance Regularized Two-Layer Level Set Approach,” in Medical Image Computing and Computer-Assisted Intervention – MICCAI 2013, K. Mori, I.
Sakuma, Y. Sato, C. Barillot, and N. Navab, Eds. Springer Berlin Heidelberg, 2013, pp.
477–484.
[36] T. A. Ngo and G. Carneiro, “Left ventricle segmentation from cardiac MRI combining level set methods with deep belief networks,” in 2013 20th IEEE International Conference on Image Processing (ICIP), 2013, pp. 695–699.
[37] C. Constantinides, Y. Chenoune, N. Kachenoura, E. Roullot, E. Mousseaux, A.
Herment, and F. Frouin, “Semi-automated cardiac segmentation on cine magnetic resonance images using GVF-Snake deformable models,” MIDAS J. - Card. MR Left Ventricle Segmentation Chall., 2009.
[38] L. Cordero-Grande, G. Vegas-Sánchez-Ferrero, P. Casaseca-de-la-Higuera, J.
Alberto San-Román-Calvar, A. Revilla-Orodea, M. Martín-Fernández, and C. Alberola-López, “Unsupervised 4D myocardium segmentation with a Markov Random Field based deformable model,” Med. Image Anal., vol. 15, no. 3, pp. 283–301, Jun. 2011.