Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Medical Biophysics

Supervisor

Jean Théberge

2nd Supervisor

R Terry Thompson

Co-Supervisor

Abstract

According to the World Health Organization (WHO) report in 2019, cardiovascular diseases (CVD) cause 52% of all illness-related deaths globally and are considered to be the second most common cause of death in Canada. CVD is also estimated to cost the Canadian economy about $21.2 billion in direct and indirect costs. With these figures, it is vital to develop the most effective and accurate methods and tools to diagnose accurately CVD and their causes. One of the promising tools for accurate diagnostic and therapeutic of CVD is the integrated Positron Emission Tomography (PET) and Magnetic Resonance Imaging (MRI) PET/MRI technology, which has successfully been used in cardiovascular imaging. The PET/MRI system provides low exposure of radioactive and ionized radiation which is advantageous over the standard technology of integrated PET/computed tomography (CT) PET/CT system. However, since the integrated PET/MRI technology was first introduced in 2010 for clinical use, its hardware attenuation correction (AC) still presents a challenge, which is crucial to achieving accurate PET quantification in cardiac imaging. Additionally, for cardiovascular PET/MRI the system still requires a higher temporal and spatial resolution radio frequency (RF) phased array for faster imaging sessions of cardiac patient, without loss of MRI image quality, while minimizing photon attenuation. This thesis introduces a novel 32-channel RF phased array, prospectively-designed for simultaneous PET/MRI cardiovascular imaging. The phased array’s MR imaging quality parameters, including, geometry factor (g-factor), noise correlation coefficient (NCC) and signal-to-noise ratio (SNR) were measured using a phantom and three healthy volunteers and the results were compared against currently used arrays.

Post-assessing the MR image quality, the array was evaluated for 511keV PET photon attenuation. The evaluation is carried out using a NEMA procedure and phantom, in which contrast recovery (CR), background variation (BV) and contrast-to-noise ratio (CNR) were measured and compared. Furthermore, the thesis presents a static radioactive source as a novel method for accurate attenuation correction (AC) of hardware (i.e. patient table) used during cardiovascular imaging. In summary, assessing both MRI and PET performances of the novel array, resulted in MRI SNR improvements of >30% at different acceleration factors (R > 2), compared to the standard array. In the meantime, the PET counts loss caused by the novel array was significantly lower (p=0.001) than those caused by the standard arrays. The novel AC method produced a hardware AC map with global counts loss of -0.7% in comparison to -4.3% as produced by the CT-based method. In conclusion, both the novel array and the hardware AC method presented here, enable the acquisition of high temporal (fast imaging session) and spatial (image quality) resolutions by the MRI system, together with accurately quantifying the PET standardized-uptake-value (SUV). The method and tools presented in this work have been evaluated for simultaneous PET/MRI cardiovascular imaging, and hence they can be effectively used to study CVD and their causes accurately in a shorter imaging time. Therefore, the improvements reported in this thesis contribute to better understand the CVD and potentially lowering the economic burden around them. The impact of these improvements is broad, since they may be applied to PET/MRI imaging of brain, prostate and other organs.

Summary for Lay Audience

Treatment of diseases is only possible if their causes and developments are understood. In medical imaging, understanding the disease is the purpose behind developing methodologies and tools that can be used during this process. Two of the imaging tools used in understanding cardiac diseases are the positron emission tomography (PET) and magnetic resonance imaging (MRI). Both have been integrated into one scanner namely PET/MRI to complement each other, which provides excellent data to study heart functions, disease causes and disease development. In hybrid PET/MRI technology, the MRI is free of radiation and provides excellent image contrast, while the PET system is used for measuring tissue functions, both of which are necessary for diagnosis.

The PET/MRI technology is fairly new (only 10 years old), and is powerful, because it reduces the study time by half especially for patients requiring imaging with both PET and MRI. However, due to the integration of the PET and MRI into one, hybrid PET/MRI, several challenges become apparent which need to be addressed. One of these challenges is the inaccurate counts of radioactivity in the PET image. This happens when the radioactivity is blocked by hardware parts which are present in the scanner during imaging. Another challenge is to obtain high-resolution MRI images and reduce the scanning time, so a tool (named a RF phased array) is required to achieve these high-resolution images and shorten the time of the imaging session for the patient. This research work addresses both the accurate counting of the radioactivity and the RF phased array. My research work provided an RF array that does not block radioactivity, while also being capable of shortening the imaging time to a tolerable level for the cardiac patient. The research work also provided a new approach to correct for inaccurate counts caused by other parts in the PET/MRI system. The author anticipates that these improvements and developments will enhance accurate diagnosis and the study of cardiac diseases using the PET/MRI platform.

Available for download on Wednesday, March 31, 2021

Share

COinS