An Automated System for 3D Segmentation of CT Angiograms - Visual Computing: Scientific Visualization and Imaging Systems - page 96

Fig. 5.1 Effect of plaque
builds up. a Shows a normal
artery with normal blood
flow. b Shows an artery with
plaque buildup [ 2 ]
In order to evaluate the development of coronary artery disease, efficient
screening procedures are urgently needed for early diagnosis and ultimately timely
treatment. From the clinician's point of view, it is essential to detect and visualise
the branches of the arteries, which may be associated to stenosis (narrowing of the
arteries). Then, potentially carrying out haemodynamic studies to establish the risk
of a patient developing a heart attack, followed by the application of appropriate
treatment based on medication or surgical procedures, such as diagnostic cardiac
catheterisation and stent implantation [ 3 , 4 ].
Recent advances in vascular imaging technology using multiple-slice Computed
Tomography (CT) provide volumetric datasets with high spatial and temporal res-
olution. This has given rise to CT becoming a frequently used and reliable imaging
modality for the early diagnosis and quantification of atherosclerosis in clinical
practice. However, clinical diagnosis of CAD by means of CT imaging is difficult
because of the huge amount of data produced in the cardiac CT scanning process
(typically, more than 300 slices of axial CT images are produced for a patient).
Interpretation of a CT study is a labour-intensive and time-consuming task even with
the assistance of semi-automatic software packages [ 5 ], as the radiologist has to track
and carefully examine each branch of the arteries, segment by segment. Therefore,
automated algorithms for segmentation of the coronaries in CT angiograms will
facilitate and enhance the accuracy and reliability of the clinical evaluation.
In the following section, we provide an overview of recent developments in the
field of the research. Next, we present the proposed framework in Sect. 5.3 . This is
followed by the presentation of results demonstrating the benefits of the proposed
approach. Finally, the conclusions and future developments of this research are
presented.