Use this URL to cite or link to this record in EThOS:
Title: Interventional vascular enhancement using digital tomosynthesis facilitated by 2D-3D registration
Author: Alhrishy, Mazen
ISNI:       0000 0004 5990 8427
Awarding Body: King's College London
Current Institution: King's College London (University of London)
Date of Award: 2016
Availability of Full Text:
Access from EThOS:
Access from Institution:
Endovascular image-guided surgery (EIGS) using C-arm fluoroscopy is being per- formed for more and more complex procedures with longer imaging times. An example of such complex procedures is fenestrated endovascular aneurysm repair (EVAR), which involves delivering bespoke stent-grafts through vasculature into the aneurismal area using low dose images. However, X-ray is much better at visualizing interventional devices and dense structures compared to vasculature. To enhance vascular visualisation, iodinated contrast medium (ICM) is injected and a significantly higher X-ray dose than standard low dose images is employed to produce a digital subtraction angiography (DSA) image. ICM usage is essential but nephrotoxic and can cause renal failure. DSA is also often a major contributor to the overall patient radiation dose. Furthermore, a DSA image is only valid for the current interventional view and not the new view once the C-arm is moved. The required high accuracy, the lengthy imaging time, and the large volumes of ICM used (with the associated radiation dose and high risks of renal failure) during fenestrated EVAR, have motivated this thesis to look into novel methods to enhance vasculature, and reduce ICM usage during complex EIGS procedures, while maintaining the clinical work-flow. My first novel method proposes using an established two-dimensional (2D)- three-dimensional (3D) rigid registration system to facilitate improved interven- tional DTS (iDTS) reconstruction, using standard hardware, and with no ICM injection. This is achieved by a small angle C-arm sweep (e.g. 40) to acquire intraoperative fluoroscopy images, which are then registered to the preoperative computed tomography (CT) volume. The proposed method automatically recon- structs patient-anatomy-specific images and removes clutter resulting from bony anatomy. Experiments were carried out using one phantom and 4 clinical datasets. Phantom results showed a 3419% signal difference to noise ratio (SDNR) improvement compared to standard fluoroscopy images. Patient results showed that the method enabled visualization of clinically relevant features: the outline of the aorta and some aortic calcifications, without the injection of any ICM, and with much reduced radiation dose compared to DSA imaging. The second method is an extension to the first method to enable much smaller vascular structures, such as the renal ostia, to be enhanced. This is achieved by injecting a much smaller ICM volume than is usually used for standard DSA imaging (e.g. 10-30%) during the C-arm sweep. Experiments were carried out using CT-based synthetic fluoroscopy images, to which simulated contrast was added in different amounts before deboned iDTS (de-iDTS) reconstruction. Numerical results showed that the reconstructed de-iDTS images with simulated contrast had better SDNR values compared to the synthetic DSA image with motion and noise artefacts. Reconstructed de-iDTS images showed that the renal ostia can be clearly seen after adding motion and noise when using 30% simulated ICM. The third novel method proposes using the mentioned 2D-3D registration system to facilitate remapping a DSA image from one view to another, after the C-arm is moved. This is achieved by registering the two image views to the preoperative CT volume, which allows repeated ICM-free DSA imaging. Experiments were carried out using 9 clinical datasets. Numerical results showed an overall averaged remapping accuracy error of 2.73 mm, with 7 patients scoring averaged errors 3 mm. In addition, the overall averaged error was found to increase by 163% when using the 2D-3D overlay method, which was found to be statistically significant (p < 0.01). In summary, I have developed novel imaging methods for vascular enhancement during interventional fluoroscopy using standard hardware. My novel methods have shown: 1) the potential to provide additional intraoperative information, which cannot be provided by the preoperative CT volume alone, such as the deformed aorta position with respect to interventional devices, and 2) the potential to enable a reduction in overall ICM usage, and radiation dose, while maintaining the clinical work flow. I propose that these methods could find a role alongside DSA imaging, replacing DSA imaging where appropriate, while using DSA for critical points in the procedure. This is particularly beneficial for patients with renal insufficiency and/or patients at high risk of radiation adverse response. Because my novel method of iDTS can be directly employed on any standard fluoroscopy system, in addition to vascular enhancement during EIGS, the method could be potentially used to enhance structures in other image-guided surgery (IGS) applications.
Supervisor: King, Andrew Peter ; Penney, Graeme Patrick Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available