Information Technology Reference
In-Depth Information
Fig. 28 The shape embedding and
final segmentation results are shown in 3D views. a A CT data
is shown in the sagittal axis (without the re
nement). b The initial location of the shape models.
2D shape models are propagated in z-axis to form 3D models. The blue color (outer volume)
shows the variability region, whereas the yellow color (inner volume) represents the object region.
c The shape model after registration. d The
final segmentation results using the three models
used as a reference to align any given VB later to use the proposed shape model.
This alignment approach is similar to the method presented in the previous method.
Finally, a
is generated, which its slices are shown
in Fig.
30
. Three regions in this shape model: white color represents
“
shape volume
” P
s
¼
O[B[V
O
(VB), black
represents
B
(its background), and gray is the variability region
V
. Figure
31
a
illustrates a 3D view of the VB and its variability region.
We use a distance probabilistic model, to model variability region
i.e., the 3D
shape variations. A normal distance is used to describe the VB and its background
in the variability region in the distance probabilistic model as follows.
V
d
p
¼
c
2
C
OV
k
min
p
c
k;
ð
30
Þ
to the organ/variability
surface C
OV
. Figure
31
b, c shows iso-surfaces for C
OV
, where an iso-surface C
d
p
is
a set of voxels located at equal distance d
p
from C
OV
. Assuming each iso-surface
Equation (
30
) represents the distance from a voxel p
2V
