Biomedical Engineering Reference
In-Depth Information
of tag lines in both the LV and RV could provide for more accurate radial strains,
although in that case one would be faced with increased challenges in terms of
accurate tag line delineation. In addition to tagging density considerations, one
must also be cautious in ensuring the fidelity of the MR images. Potential prob-
lems include motion artifacts (e.g., related to breath-holds) and misalignment of
the short- and long-axis images due to variabilities in heart motion. These are not
issues restricted to this study, but concern all cardiac MR applications. In general,
however, with increased acquisition speed, particularly in the use of parallel imag-
ing techniques, these issues may become less of a concern. We note that our canine
studies involved manual breath-holds, and so issues related to breathing motion
artifacts were not encountered. In human studies the subjects underwent training
prior to the actual imaging session so that the breath-holds always took place at
end-inspiration and subjects were cued as to when they could breathe (duration
of all breath-holds remained the same across acquisition of all images for a given
study). In a few instances for both human and canine imaging, misalignments were
observed, and these were resolved by ensuring consistency between corresponding
myocardial regions in short- and long-axis images.
Alternative and complementary techniques for tagged MRI have been devel-
oped for greater spatial resolution such as DENSE [36] and HARP [37, 38], from
which dense displacement information could be inferred. The NURBS method-
ology should prove helpful in spatiotemporal filtering of data from these alternate
motion tracking methods and can lead to accurate estimates of 4D myocardial
kinematics.
In addition to further development of the technical aspects of our research,
current applications are directed to a variety of disease pathophysiologies. We
have looked at patient data exhibiting ischemic heart disease and LV hypertrophy,
and plan to apply our technique to RV hypertrophic models.
8.
REFERENCES
1.
S. Masood, Yang G-Z, Pennell DJ, Firmin DN. 2000. Investigating intrinsic myocardial me-
chanics: the role of MR tagging, velocity phase mapping, and diffusion imaging.
JMagn
Reson Imaging
12
:873-883.
2.
Amini AA, Prince JL, eds. 2001.
Measurement of cardiac deformations from mri: physical
and mathematical models
. Dordrecht: Kluwer Academic
3.
Frangi AF, Niessen WJ, Viergever MA. 2001. Three-dimensional modeling for functional
analysis of cardiac images: a review.
IEEE Trans Med Imaging
20
(1):2-25.
4.
Haber I, Metaxas DN. 2000. Three-dimensional motion reconstruction and analysis of the
right ventricle using tagged MRI.
Med Image Anal
4
:335-355.
5.
Klein SS, Graham TP, Lorenz CH. 1998. Noninvasive delineation of normal right ventricular
contractile motion with magnetic resonance imaging myocardial tagging.
Ann Biomed Eng
26
:756-763.
6.
Naito H, Arisawa J, Harada K, Yamagami H, Kozuka T, Tamura S. 1995. Assessment of right
ventricular regional contraction and comparison with the left ventricle in normal humans: a
cine magnetic resonance study with presaturation myocardial tagging.
Br Heart J
74
:186-191.
