3D IMAGE RECONSTRUCTION TECHNIQUES FOR CONE BEAM COMPUTED TOMOGRAPHY

Abstract

CBCT; Cone-beam; Projections; Analytical reconstruction methods; Iterative reconstruction methods. Summary:

Computed Tomography (CT) is a widely used imaging technique in medical diagnosis. It relies on collecting data about an object from multiple views representing its projections and subsequently using these data to construct an image. Among the latest advances in CT is the utilization of cone-beam geometry of x-ray projections is used instead of the usual planar fan beam. This technique is called Cone Beam CT (CBCT) and promises even faster yet safer 3D imaging compared to previous CT technologies. These advantages come at the expense of a more challenging reconstruction problem that is still an active research area to optimize for speed and image quality. Present reconstruction methods are classified into two main categories, namely; analytical and iterative methods. In this thesis, we study the performance of the leading reconstruction methods from both categories and investigate modifications in their parameter selection. We also compared their quality, speed and dose quantity of x-ray radiation using analytical phantoms. The results indicate that the reconstructed volumes produced from the iterative methods provide better quality but at the cost of a higher computational complexity. We develop a parallel computing version of the algorithms and show that the computation time can be improved using multi-core processing. The results of this work provide a framework for optimizing the quality, computation time, projection data required and hence the patient dose.