Page 12 - Artificial Intelligence for Computational Modeling of the Heart

List of figures xi

Fig. 2.15 Computational performance of the LBM-EP algorithm on different
architectures: single processor, multicore processing on the CPU
and on the graphical processing unit. 57
Fig. 2.16 Illustration of the spontaneous activation points. 58
Fig. 2.17 Graphical representation of the modeling approach for high-speed
conducting tissue. The lattice nodes of the Cartesian grid are
shown in the background, colored by the local value of the level
set. The triangulated surface represents the endocardium. For one
of the lattice nodes, the sub-grid defined in the voxel is visualized,
each point in the sub-grid being colored by the value of the level
set. In this example we considered a threshold h = 0.1 mm. For
visualization purposes, the color bar has been scaled to the
interval [−0.2,0.2] mm. 60
Fig. 2.18 Overview of the workflow for computational modeling of
patient-specific ECG. 63
Fig. 2.19 Left Panel: general principle of geometry matrix definition using
the example of P HB (observation points on the heart and torso as
integration surface), see text for details. Right panel: actual
implementation principle of P HB calculation using triangular
mesh representation of body surface. Different triangles on the
torso are represented using different shades of red (mid gray in
print version). 65
Fig. 2.20 Variation of the active contraction stress τ c (t) (in blue (dark gray
in print version)) with respect to the electrical command function
u(t) (in red (mid gray in print version)) controlled by the cardiac
electrophysiology model. 69
Fig. 2.21 Cardiac pressure-flow system with pressure driven valves that
modulate the interactions between the ventricles, arteries and
atria. The remote pressures can be set independently to
physiological values or can be connected as part of a whole body
circulation system. 70
Fig. 2.22 The effect of arteries and atria on the ventricles is modeled by
attaching springs to the valve plane. 72
Fig. 2.23 The pericardium region is defined by the epicardium at end
diastole. Left panel: the valves are closed automatically to define
an inner and outer region. A distance map is built (middle panel),
to identify regions in which the epicardium is allowed to freely
slide along the pericardial bag (authorized region in white,
epicardium in black). As soon as an epicardial node goes outside
the authorized region, a force along the gradient map (right panel)
is applied to bring that node back. 72
Fig. 2.24 A mode of simple shear defined with respect to the fiber, sheet
and normal axes. The first letter in (sf) stands for the normal
vector to the face that is subject to shear, the second letter
denotes the direction of shear. (Source: [120].) 75
Fig. 2.25 Simple shear tests on a finite sample of myocardial tissue. Circles
represent experimental data as provided by [112]. Plain curves
represent computational results. Each label summarizes the test
as follows: the first letter stands for the normal vector to the face