Abstract
We propose a new method for myocardial material point tracking on cine magnetic resonance images which makes use of a multiresolution Markov random field of discrete 3D deformations defined on a tetrahedral grid of the myocardial volume. A block-matching procedure is combined with biomechanics-based and smoothness assumptions on the resulting deformations. Quantitative results and visual illustration about the cardiac mechanics parameters obtained are provided on a sample of 7 patients affected by different cardiomyopathies.
© 2013 Optical Society of America
PDF ArticleMore Like This
Michael J. Swain and Lambert E. Wixson
WC2 Image Understanding and Machine Vision (IUMV) 1989
Chi-hsin Wu and Peter C. Doerschuk
RWC2 Signal Recovery and Synthesis (SRS) 1995
Takeo Minamikawa, Tsunehisa Yamamoto, Yoshinori Harada, Yoshihisa Yamaoka, Hideo Tanaka, Hitoshi Yaku, and Tetsuro Takamatsu
cl_p_11 The European Conference on Lasers and Electro-Optics (CLEO/Europe) 2019