Phase contrast x-ray velocimetry of the lungs: from the synchrotron to the laboratory

2017-05-15T07:26:30Z (GMT) by Murrie, Rhiannon Paige
In this thesis the imaging of the lungs via propagation-based phase contrast x-ray imaging (PB-PCXI) and x-ray velocimetry (XV) is studied, both on a synchrotron and laboratorybased x-ray system. Current clinical imaging techniques, such as magnetic resonance imaging (MRI), positron emission tomography (PET) or computed tomography (CT), lack either the spatial resolution to image the lungs in fine detail and/or the temporal resolution to image the lungs fast enough to avoid motion blur whilst breathing. PB-PCXI and XV of small animal lungs at high resolution has previously been demonstrated on the SPring-8 synchrotron in Japan, and this thesis looks at the feasibility of performing these techniques both at the Australian Synchrotron and on a laboratory x-ray system, with the ultimate goal of clinical implementation in mind. Firstly, we investigate the feasibility of performing both PB-PCXI and XV of small animal lungs, in vivo, on the Imaging and Medical Beamline (IMBL) of the Australian Synchrotron. This study examines the optimal experimental parameters for PB-PCXI of the lungs, such as sample-to-detector propagation distance, primary slit width and monochromator crystal bend in the two experimental hutches of the IMBL designed for PCXI, and ultimately looks at whether performing PB-PCXI and XV of the lungs is viable on the IMBL. We then demonstrate live animal XV of the lungs on the IMBL, the first live animal images to be acquired at this facility. The difficulties of performing XV on the live lung images acquired are discussed and a range of different analysis methods presented. High resolution micro-CT of the lungs and airways is also demonstrated. Finally, we explore the optimal imaging rates and sequences for performing XV of the lungs on both a synchrotron and a laboratory source via simulation. The simulation model developed is based on experimental data acquired on each of the respective sources, and allowed a comprehensive investigation of the data space to be undertaken. The exposure time ratio (exposure time/time between exposures) is identified as an important parameter in XV and the optimal value is calculated, along with the optimum exposure times for imaging on both a synchrotron and a laboratory source. The limitations of rotational motion in XV are also identified and discussed. It is hoped that the research undertaken in this thesis will assist in obtaining an increased knowledge of lung dynamics and function, which in turn will increase our ability to diagnose and treat chronic lung diseases.