Combining Spectral and Active Shape Methods to Track Tagged MRI

Tagged magnetic resonance is a very usefull and unique tool that provides a complete local and global knowledge of the left ventricle (LV) motion. In this article we introduce a method capable of tracking and segmenting the LV. Spectral methods are applied in order to obtain the so called HARP images which encode information about movement and are the base for LV point-tracking. For segmentation we use Active Shapes (ASM) that model LV shape variation in order to overcome possible local misplacements of the boundary. We finally show experiments on both synthetic and real data which appear to be very promising.

[1]  Jerry L. Prince,et al.  Imaging heart motion using harmonic phase MRI , 2000, IEEE Transactions on Medical Imaging.

[2]  Dmitry B. Goldgof,et al.  Automatic tracking of SPAMM grid and the estimation of deformation parameters from cardiac MR images , 1994, IEEE Trans. Medical Imaging.

[3]  W. O'Dell,et al.  Three-dimensional myocardial deformations: calculation with displacement field fitting to tagged MR images. , 1995, Radiology.

[4]  Jerry L. Prince,et al.  Tag and contour detection in tagged MR images of the left ventricle , 1994, IEEE Trans. Medical Imaging.

[5]  Jerry L Prince,et al.  Cardiac motion tracking using CINE harmonic phase (HARP) magnetic resonance imaging , 1999, Magnetic resonance in medicine.

[6]  Elliot R. McVeigh,et al.  Four-dimensional B-spline-based motion analysis of tagged cardiac MR images , 1999, Medical Imaging.

[7]  Jerry L. Prince,et al.  Image processing and analysis in tagged cardiac MRI , 2000 .

[8]  A. Young,et al.  Three-dimensional motion and deformation of the heart wall: estimation with spatial modulation of magnetization--a model-based approach. , 1992, Radiology.

[9]  Jerry L Prince,et al.  Cardiac motion simulator for tagged MRI , 1996, Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis.

[10]  Jerry L. Prince,et al.  Reconstruction of 3-D left ventricular motion from planar tagged cardiac MR images: an estimation theoretic approach , 1995, IEEE Trans. Medical Imaging.

[11]  Timothy F. Cootes,et al.  Active Shape Models-Their Training and Application , 1995, Comput. Vis. Image Underst..

[12]  P. Radeva,et al.  Deformable B-solids and implicit snakes for localization and tracking of SPAMM MRI-data , 1996, Proceedings of the Workshop on Mathematical Methods in Biomedical Image Analysis.