Biomedical Engineering Reference
In-Depth Information
Chapter 6
Inverse Consistent Image Registration
G. E. Christensen
1
6.1
Introduction
Image registration has many uses in medicine such as multimodality fusion,
image segmentation, deformable atlas registration, functional brain mapp-
ing, image guided surgery, and characterization of normal vs. abnormal anatom-
ical shape and variation. The fundamental assumption in each of these
applications is that image registration can be used to define a meaningful corre-
spondence mapping between anatomical images collected from imaging devices
such as CT, MRI, cryosectioning, etc. It is often assumed that this correspon-
dence mapping or transformation is one-to-one, i.e., each point in image
T
is
mapped to only one point in image
S
and vice versa. A fundamental problem
with a large class of image registration techniques is that the estimated transfor-
mation from image
T
to
S
does not equal the inverse of the estimated transform
from
S
to
T
. This inconsistency is a result of the matching criteria's inability
to uniquely describe the correspondences between two images. Inverse con-
sistent registration seeks to overcome this limitation by jointly estimating the
transformation from
T
to
S
and from
S
to
T
while minimizing the correspondence
inconsistencies between the forward and reverse transformations. Forward and
reverse transformations that are inverses of each other are defined to be inverse
consistent with each other. The inverse consistency error is a measure of the
difference between the forward transformation and the inverse of the reverse
transformation, and vica versa.
