This paper presents a framework for tracking left ventricular (LV) endocardium through 2D echocardiography image sequence. The framework is based on fusion of biomechanical (BM) model of the heart with the parametric deformable model. The BM model constitutive equation consists of passive and active strain energy functions. The deformations of the LV are obtained by solving the constitutive equations using ABAQUS FEM in each frame in the cardiac cycle. The strain energy functions are defined in two user subroutines for active and passive phases. Average fusion technique is used to fuse the BM and deformable model contours. Experimental results are conducted to verify the detected contours and the results are evaluated by comparing themto a created gold standard. The results and the evaluation proved that the framework has the tremendous potential to track and segment the LV through the whole cardiac cycle. 1. Introduction Echocardiography is an important imaging modality that enables the cardiologist to evaluate the structure and functions of the heart. Because of noninvasive characteristics, low cost, and being nonionizing radiation, echocardiography has been largely applied in the evaluation of cardiac function. One of the most important applications of echocardiography is in determining systolic and diastolic ventricular volumes of the patient, both of which are used to calculate the left ventricular ejection fraction, muscle contraction ratio of cardiac cavities, local ejection fraction, myocardial thickness, and the ventricle mass [1]. To calculate the above-mentioned parameters, the cardiac muscle contour on the echocardiography image needs to be identified. The border detection process simplifies image analysis and greatly reduces the amount of data which needs to be processed, while preserving the structural information about the contours of the object under study [2]. However, in clinical practice, this task still relies on manual outlining. Manual outlining of these borders is slow, time consuming, and tedious task. Moreover, the resulting outlines vary between different observers and suffer from a subjective bias [3]. Automatic LV border detection and tracking over the cardiac cycle in echocardiographic image sequences remain open and a challenging problem due to many difficulties related to the heart and its dynamics and other difficulties related to the echocardiography ultrasound machine. Echocardiography has a poor image quality and resolution with various image artifacts like speckle, shadowing, and side lobes [3]. The images of
References
[1]
S. Chan and G. Sainarayanan, “Fuzzy-based boundary enhancement for echocardiogram using local image characteristics,” Malaysian Journal of Computer Science, vol. 19, no. 2, pp. 151–159, 2006.
[2]
H. Ketout, J. Gu, and G. Horne, “Improved dempster and shafer theory to fuse fuzzy inference system, neural networks and CNN endocardial edge detection,” The Mediterranean Journal of Electronics and Communications, vol. 7, no. 3, pp. 267–275, 2011.
[3]
G. Sainarayanan, N. Mal Murugan, and S. Chan, “A novel method for echocardiogram boundary detection using adaptive neuro-fuzzy systems,” in Proceedings of the International Conference on Computational Intelligence and Multimedia Applications (ICCIMA '07), vol. 3, pp. 415–419, December 2007.
[4]
J. G. Bosch, S. C. Mitchell, B. P. F. Lelieveldt et al., “Automatic segmentation of echocardiographic sequences by active appearance motion models,” IEEE Transactions on Medical Imaging, vol. 21, no. 11, pp. 1374–1383, 2002.
[5]
L. F. Zhang and E. A. Geiser, “An effective algorithm for extracting serial endocardial borders from 2-dimensional echocardiograms,” IEEE Transactions on Biomedical Engineering, vol. 31, no. 6, pp. 441–447, 1984.
[6]
P. M. F. Nielsen, I. J. Le Grice, B. H. Smaill, and P. J. Hunter, “Mathematical model of geometry and fibrous structure of the heart,” American Journal of Physiology: Heart and Circulatory Physiology, vol. 260, no. 4, pp. H1365–H1378, 1991.
[7]
M. P. Nash and P. J. Hunter, “Computational mechanics of the heart,” Journal of Elasticity, vol. 61, no. 1–3, pp. 113–141, 2000.
[8]
D. H. S. Lin and F. C. P. Yin, “A multiaxial constitutive law for mammalian left ventricular myocardium in steady-state barium contracture or tetanus,” Journal of Biomechanical Engineering, vol. 120, no. 4, pp. 504–517, 1998.
[9]
V. Chalana and Y. Kim, “A methodology for evaluation of boundary detection algorithms on medical images,” IEEE Transactions on Medical Imaging, vol. 16, no. 5, pp. 642–652, 1997.
[10]
C. Li, C. Xu, C. Gui, and M. D. Fox, “Level set evolution without re-initialization: a new variational formulation,” in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR '05), vol. 1, pp. 430–436, June 2005.
[11]
A. Blake and M. Isard, Active Contours, 1st edition, 2000.
[12]
A. Blake, M. Isard, and D. Reynard, “Learning to track the visual motion of contours,” Artificial Intelligence, vol. 78, no. 1-2, pp. 179–212, 1995.
[13]
F. Dorri, P. F. Niederer, and P. P. Lunkenheimer, “A finite element model of the human left ventricular systole,” Computer Methods in Biomechanics and Biomedical Engineering, vol. 9, no. 5, pp. 319–341, 2006.
[14]
Dassault Systèmes Simulia Corp., Providence, RI, USA, Abaqus User Subroutines Reference Manual, 2010.
[15]
Dassault Systèmes Simulia Corp., Providence, RI, USA, ABAQUS THEORY MANUAL, 2010.
Dassault Systèmes Simulia Corp., Providence, RI, USA, ABAQUS Analysis User's Manual, Vol. 1 : Introduction, Spatial Modeling, Execution and Output.
[18]
M. Kass, A. Witkin, and D. Terzopoulos, “Snakes: active contour models,” International Journal of Computer Vision, vol. 1, no. 4, pp. 321–331, 1988.
[19]
I. M. Anderson and J. C. Bezdek, “Curvature and tangential deflection of discrete arcs: a theory based on the commutator of scatter matrix Pairs and its application to vertex detection in planar shape data,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 6, no. 1, pp. 27–40, 1984.
