Devon Richtsmeier

Topic

Exploration of applications of photon-counting detector computed tomography (CT) using a table-top CT system

Department of Physics and Astronomy

Date & location

• Monday, April 8, 2024
• 10:00 A.M.
• Clearihue Building, Room B007

Examining Committee

Supervisory Committee

• Dr. Magdalena Bazalova-Carter, Department of Physics and Astronomy, University of Victoria (Co-Supervisor)
• Dr. Matthew Moffitt, Department of Chemistry, UVic (Co-Supervisor)
• Dr. Derek Wells, Department of Physics and Astronomy, UVic (Member)

External Examiner

• Dr. Scott Hsieh, Department of Radiology, Mayo Clinic

Chair of Oral Examination

• Dr. Yin-Man Lam, Department of Anthropology, UVic

Abstract

Computed Tomography (CT) is an essential diagnostic tool in healthcare, widely used for various applications including cancer detection, vascular disease evaluations, and radiation therapy planning. Recent advancements in photon-counting detector (PCD) technology have led to the development of photon-counting detector CT (PCD-CT), a promising innovation offering high spatial resolution and superior contrast-to-noise ratios compared to traditional CT. PCD-CT excels in detecting and characterizing small structures in various body parts, enhancing tissue differentiation, and material decomposition, thus potentially improving disease diagnosis and radiotherapy treatment planning. This research explores the applications of PCD-CT using a bench-top system, offering insights into its potential benefits over conventional CT systems. Despite its limitations in fully representing a clinical CT system, the bench-top model provides flexibility in assessing clinically useful features. This dissertation investigates four key applications of PCD-CT: material decomposition, multi-contrast imaging, metal artifact reduction, and high-resolution imaging.

We investigated the material decomposition capabilities of our bench-top PCD-CT scanner using a dual-energy CT (DECT) method for extracting effective atomic number (Z_eff) and relative electron density (ρ_e) of tissues. We demonstrated that the method with PCD-CT was more accurate in extracting Z_eff and ρ_e for a set of electron density phantom materials with known Z_eff and ρ_e than the method with DECT. In addition, four tissue types were correctly identified in an ex-vivo tissue sample and an injected gold contrast agent was separated from the other four tissue types using K-edge subtraction imaging.

Multi-contrast imaging was demonstrated in a phantom model with four contrast agents: gadolinium, dysprosium, lutetium, and gold. The four contrast agents were inserted into the same cylindrical phantom and imaged in one scan. Using K-edge subtraction, we were able to demonstrate complete separation and accurate quantification of the four contrast agents, even of gadolinium and dysprosium, which have K-edge energies of 50.2 keV and 53.8 keV, respectively. Additionally, we optimized the acquisition parameters for the various contrast agents.

We also developed a novel metal artifact reduction (MAR) method using PCD-CT. As metal attenuates fewer higher energy x-rays than low energy x-rays, we showed that the high-energy range of 100–110 keV demonstrated fewer metal artifacts. The high energy range is separable from the other x-ray data with the PCD. With this in mind, we developed trace replacement metal artifact reduction (TRMAR). The metal traces in the corrupted conventional CT sinogram space are replaced with the high-energy trace data from the 100–110 keV range. With this, we maintained the contrast and image quality of the full spectrum conventional CT image and also kept the reduced metal artifacts from the high energy data.

Finally, we demonstrated the high spatial resolution of our PCD-CT system by imaging coronary artery stents and comparing the same stents imaged with two conventional CT scanners. PCD-CT demonstrated more accurate measurement of the stent strut, or wire, width, stent lumen diameter, and lumen CT number compared to conventional CT. In addition, this led to more accurate 3D representations of the stents. The higher accuracy of strut width and stent visualization is due to the higher spatial resolution of the PCD-CT system and the reduced metal and blooming artifacts it offers over conventional CT.

Each of these applications demonstrates the significant potential of PCD-CT in enhancing medical diagnostics and treatment, particularly in cardiovascular imaging, highlighting its diverse contributions to the field of medical imaging.