A Survey of Three-Dimensional Modeling Techniques for Quantitative Functional Analysis of Cardiac Images - Advance Image Processing in Magnetic Resonance Imaging - page 278

relatively little effort. Phase-contrast imaging has also been applied to investigate
regional myocardial function, in a similar way as MR tagging. By acquiring multiple
short-axis images with different orthogonal phase-encoding directions, three veloc-
ity components of a material point inside the myocardium can be computed, pro-
viding a dense displacement field for the myocardium with a higher resolution than
tagging data. Analogous to tagging analysis, the displacement field can serve as
input to a continuum model, yielding estimates for myocardial stress and strain.
Compared to MSCT, cardiac MRI provides similar data, the additional advan-
tages being the nonionizing nature of the modality, its versatility, and superiority
in flow and perfusion imaging. MSCT, on the other hand, enables a significantly
faster acquisition ( 20 sec), and thus enables a higher patient throughput and the

simultaneous study of coronary and ventricular function.

±

9.3 CLASSICAL DESCRIPTORS OF CARDIAC

FUNCTION

Development of models of the cardiac chambers has emerged from different dis-
ciplines and with various goals. Cardiac models have been used for deriving func-
tional information, for visualization and animation, for simulation and planning of
surgical interventions, and for mesh generation for FE analysis.
This survey will be confined to the application of modeling techniques for obtain-
ing classical functional analysis. Classical functional analysis can be divided into global
functional analysis (Subsection 9.3.1) and motion and deformation analysis (Subsec-
tion 9.3.2), from which the most clinically relevant parameters can be obtained. *
Model-based methods also allow one to derive new descriptors of cardiac shape
and motion. Such advanced descriptors have been mainly presented in the technical
literature and their clinical relevance is yet to be assessed. Without pretending to be

exhaustive, Appendix A summarizes a number of nonclassical shape and motion

descriptors that demonstrate the extra possibilities provided by some of the advanced
methods.

9.3.1 G LOBAL F UNCTIONAL A NALYSIS

Weber and Hawthorne [58] proposed a classification of cardiac indices according
to their intrinsic dimensionality: linear, surface, or volumetric descriptors. Linear
parameters have been used intensively in the past because they can easily be derived
from 2-D imaging techniques such as 2DE and x-ray angiocardiography.* However,
these techniques assume an “idealized” geometry of the LV and strongly depend
on external or internal reference and coordinate systems. Besides total ventricular
wall area, other surface indices based on curvature and derived parameters have
been investigated from 2-D studies [59-62]. More recently, many image processing

* Such parameters are, for instance, left-ventricular internal dimension (LVID), relative wall thickness
(RWT), and estimates of fractional shortening of the cardiac fibers (%∆ D ) and their velocity ( V cf ).
For a detailed analysis of these parameters refer to Vuille and Weyman [ 14 ].