Biomedical Engineering Reference
In-Depth Information
The most commonly clinically utilized techniques for the assessment of my-
ocardial regional wall motion and deformation of the myocardium are echocar-
diography and tagged MRI. LV wall function is typically assessed using 2-D
Doppler echocardiography [77-82] through the interrogation of the LV from
various views to obtain an estimate of the 3-D segmental wall motion. How-
ever, these measurements are not three-dimensional in nature. Furthermore,
echocardiography is limited to certain acquisition windows.
The most widely used approach for determining ventricular deformation is
MR tagging [83-88]. MR tagging techniques rely on local perturbation of the
magnetization of the myocardium with selective radio-frequency (RF) satura-
tion to produce multiple, thin, tag planes during diastole. The resulting mag-
netization reference grid persists for up to 400 ms and is convected with the
myocardium as it deforms. The tags provide fiducial points from which the
strain can be calculated [85, 89]. The primary strength of MRI tagging is that
noninvasive
in vivo
strain measurements are possible [85, 89]. It is effective
for tracking fast, repeated motions in 3-D. There are, however, limitations in
the use of tagged MRI for cardiac imaging. The measured displacement at a
given tag point contains only unidirectional information; in order to track the
full 3-D motion, these data have to be combined with information from other or-
thogonal tag-sets over all time frames [76]. The technique's spatial resolution is
coarser than the MRI acquisition matrix. Furthermore, the use of tags increases
the acquisition time for the patient compared to standard cine-MRI, although
improvements in acquisition speed have reduced the time necessary for image
acquisition.
Sinusas
et al
. have developed a method to determine the strain distributions
of the left ventricle using untagged MRI [90]. The system is a shape-based ap-
proach for quantifying regional myocardial deformations. The shape properties
of the endo-and epicardial surfaces are used to derive 3-D trajectories, which are
in turn used to deform a finite element mesh of the myocardium. The approach
requires a segmentation of the myocardial surfaces in each 3-D image data set
to derive the surface displacements.
Our long-term goal is to use Hyperelastic Warping to determine the strain
distribution in the normal left ventricle. These data will be compared with the left
ventricular function due to the pathologies described above. Toward this end,
the initial validation of the use of Hyperelastic Warping with cardiac cine-MRI
images is described.
