Information Technology Reference
In-Depth Information
Graphical Model-Based Vertebra
Identication from X-Ray Image(s)
Xiao Dong and Guoyan Zheng
Abstract Automated identi
cation of vertebrae from X-ray image(s) is an
important step for various medical image computing tasks such as 2D/3D rigid and
non-rigid registration. In this chapter we present a graphical model-based solution
for automated vertebra identi
cation from X-ray image(s). Our solution does not
ask for a training process using training data and has the capability to automatically
determine the number of vertebrae visible in the image(s). This is achieved
by combining a graphical model-based maximum a posterior probability (MAP)
estimate with a mean-shift based clustering. Experiments conducted on simulated
X-ray images as well as on a low-dose low quality X-ray spinal image of a scoliotic
patient veri
ed its performance.
1 Introduction
Several studies have shown automated identi
cation of vertebral bodies from
medical image(s) is important for medical image processing tasks such as seg-
mentation, registration, reconstruction and intervertebral disc identi
cation. Due to
the complexity of the spinal structure, simple feature (for example, landmarks or
edges) based solutions are not reliable and researchers are paying more attention to
graphical model-based solutions [
1
5
]. The reason to use graphical models lies in
-
the following observations:
1. Human spine is a multi-component object with a stable anatomical structure. It
is preferable to explore those structural constraints among components to
achieve a joint identi
cation of vertebral bodies or intervertebral discs rather
than dealing with them independently.
