Model-based reconstruction algorithms for dynamic X-ray CT
13 December 2017
Campus Drie Eiken, R0.19 - Universiteitsplein 1 - 2610 Antwerpen-Wilrijk (route: UAntwerpen, Campus Drie Eiken
Organization / co-organization:
Department of Physics
Vincent Van Nieuwenhove
Jan Sijbers & Jan De Beenhouwer
Public defence of the PhD thesis of Mr. Vincent Van Nieuwenhove - Faculty of Science - Department of Physics
Computer tomography (CT) is a non-invasive imaging technique. It allows visualizing the inside of an object starting from a number of x-ray images acquired from different angles around the object.
Unfortunately, several dynamic processes can disturb the imaging process such as incoming X-ray beam variation or object deformation during the CT scan. These dynamic processes are typically not taken into account causing artifacts in the reconstructed images, as such complicating further image analysis.
In this thesis, a number of improved CT image reconstruction algorithms are proposed. These algorithms aim to obtain accurate images even in the presence of unwanted dynamic processes. In the first part of the thesis, a method that models and corrects varying incoming beam intensity is introduced. The second part introduces two algorithms that take the dynamics of the object into account.
The proposed techniques have applications in both clinical and industrial applications such as improved image quality in the presence of patient motion and more accurate visualization of sample compression tests.