Analyser based imaging applied to visualising cartilage and the development of technique specific phantoms

This thesis investigates the application and quantification of the Analyser Based Imaging (AB!) techniques Diffraction Enhanced imaging (DEI) and Multiple Image Radiography (MIR). There were two aims for this work; first, to develop phantoms capable of characterising the images produced by the different sources of contrast inherent to the ABI methods; and second, to study the application of these techniques to imaging cartilage. Phantoms for the refraction and absorption contrast were developed and imaged using DEI and MIR. Analysis of the absorption images of the absorption phantom showed that the results for both techniques were in line with those predicted by theory. The refraction images of the refraction phantom were as expected. However, further analysis demonstrated the necessity for accuracy in both positioning the phantom in the beam and in positioning the analyser crystal on the rocking curve. Preliminary investigations into the use of colloidal solution in an Ultra-Small Angle X ray Scattering (USAXS) phantom showed that it is plausible that range' of colloidal solution concentrations could be used to create' a range of contrast. Signal-to-Noise Ratio area (SNRAREA) analysis was carried out for all of the phantom experiments. The results for the absorption phantom showed that the MIR images consistently gave better results than the DEI images, showing the strength of this technique for the absorption contrast, whilst the refraction phantom results highlighted discrepancies between the two ABI methods at large magnitudes of tana. Several cartilage experiments were carried out, imaging a range of samples. Cartilage could be seen in the refraction images. However, this was dependent on the size of the sample and the amount of superposition of the surrounding tissue when imaging whole joints. Signal-to-Noise Ratio line (SNRLINE) analysis was carried out for two of the samples and demonstrated a possible link between high SNRLINE values and high contrast in the Magnetic Resonance Image comparisons. Line profiles were also taken of the cartilage regions of a disarticulated sample showing the potential for detecting different cartilage layers.EThOS - Electronic Theses Online ServiceGBUnited Kingdo