Fuzzy shape based motion evaluation of left ventricle using genetic algorithm

A shape based non-rigid cardiac motion study is presented using simple fuzzy shape descriptors. The objective of this work is to evaluate the detail point wise motion trajectories from sequential contours. The shape correspondence on endocardial contour has been performed in multiple stages with well-defined, level specific curvature information. We incorporate non-uniform expansion and contraction of shape matched templates to optimize the correspondence in each level. However, final flow field evaluation is a constrained optimization problem, which results into a smooth mapping of contours. Constrained non-linear optimization with genetic algorithm has shown considerable promise in solving this problem. The results are quite consistent when correlated with the movement of implanted markers in an experimental set-up. Even though tracking contours in the reverse direction is irrelevant from a practical standpoint a good correlation between motions in either direction is observed. The algorithm has been tested over sets of 2D images to quantify the motion of left ventricle (LV) using two different imaging modalities.

[1]  P. K. Dutta,et al.  A GA based approach for boundary detection of left ventricle with echocardiographic image sequences , 2003, Image Vis. Comput..

[2]  J. Ehrhardt,et al.  Automated identification of left ventricular borders from spin-echo magnetic resonance images. Experimental and clinical feasibility studies. , 1991, Investigative radiology.

[3]  Guido Valli,et al.  A visual framework for the study of cardiac motion , 1990, [1990] Proceedings Computers in Cardiology.

[4]  James S. Duncan,et al.  Shape-based tracking of left ventricular wall motion , 1997, IEEE Transactions on Medical Imaging.

[5]  E. Zerhouni,et al.  Human heart: tagging with MR imaging--a method for noninvasive assessment of myocardial motion. , 1988, Radiology.

[6]  R. M. Mersereau,et al.  Automatic boundary detection on cardiac magnetic resonance image sequences for four dimensional visualization of the left ventricle , 1990, [1990] Proceedings of the First Conference on Visualization in Biomedical Computing.

[7]  Haiyan Wang,et al.  Geometric active deformable models in shape modeling , 2000, IEEE Trans. Image Process..

[8]  P. K. Dutta,et al.  Non-rigid cardiac motion quantification from 2D image sequences based on wavelet synthesis , 2001, Image Vis. Comput..

[9]  J. Duncan,et al.  A bending energy model for measurement of cardiac shape deformity. , 1991, IEEE transactions on medical imaging.

[10]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[11]  Songde Ma,et al.  Robust epipolar geometry estimation using genetic algorithm , 1998, Pattern Recognit. Lett..

[12]  James S. Duncan,et al.  Point-tracked quantitative analysis of left ventricular surface motion from 3-D image sequences , 2000, IEEE Transactions on Medical Imaging.

[13]  G T Meester,et al.  Quantitative assessment of regional left ventricular motion using endocardial landmarks. , 1986, Journal of the American College of Cardiology.

[14]  Bir Bhanu,et al.  Adaptive image segmentation using a genetic algorithm , 1989, IEEE Transactions on Systems, Man, and Cybernetics.

[15]  Frederic Fol Leymarie,et al.  Tracking Deformable Objects in the Plane Using an Active Contour Model , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  Nicholas Ayache,et al.  Tracking Points on Deformable Objects Using Curvature Information , 1992, ECCV.

[17]  L H Oliver,et al.  A second-generation computer-based edge detection algorithm for short-axis, two-dimensional echocardiographic images: accuracy and improvement in interobserver variability. , 1990, Journal of the American Society of Echocardiography : official publication of the American Society of Echocardiography.

[18]  G Bashein,et al.  Matched filter identification of left-ventricular endocardial borders in transesophageal echocardiograms. , 1990, IEEE transactions on medical imaging.

[19]  Yongmin Kim,et al.  A multiple active contour model for cardiac boundary detection on echocardiographic sequences , 1996, IEEE Trans. Medical Imaging.

[20]  E J Delp,et al.  Detecting left ventricular endocardial and epicardial boundaries by digital two-dimensional echocardiography. , 1988, IEEE transactions on medical imaging.

[21]  P. Anandan,et al.  A computational framework and an algorithm for the measurement of visual motion , 1987, International Journal of Computer Vision.

[22]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[23]  Suchendra M. Bhandarkar,et al.  An edge detection technique using genetic algorithm-based optimization , 1994, Pattern Recognit..

[24]  James S. Duncan,et al.  A Unified Framework to Assess Myocardial Function from 4D Images , 1995, CVRMed.

[25]  Surendra Ranganath,et al.  Contour extraction from cardiac MRI studies using snakes , 1995, IEEE Trans. Medical Imaging.

[26]  James S. Duncan,et al.  Boundary Finding with Parametrically Deformable Models , 1992, IEEE Trans. Pattern Anal. Mach. Intell..