Abstract
The scattering of photons at angles the order of 10° and less is dominated by the coherent process. Coherent scatter carries information about materials structure on the atomic or molecular scale and hence allows differentiation between low-Z materials which are of interest in biomedical applications in particular, differentiation between different types of cancer. Position-sensitive methods based on energy-dispersive X-ray diffraction (EDXRD) can be used to reconstruct images, in which each pixel carries spectral information. However, previously proposed imaging methods have the disadvantage of being time consuming. The aim of this work is to reduce the acquisition time by optimizing and testing different system geometries with the final aim is to build a system for functioning in an operating theatre environment. After a review on users of X-ray diffraction for tissue analysis, the experimental part of this work is divided into three parts. In the first one, the characteristics of an EDXRD system based on a single point detector (CdTe) and on different scatter and primary collimators geometry are discussed, using different types of samples with different thicknesses. Work also included the investigation of the effect of different scatter angles (4.5°, 6°, 7°, 8.1° and 9.6°) on the diffraction patterns. Results showed that up to the thicknesses investigated the variation in peak width was undetectable and that the effect of different angles on both peak position and peak width showed a negligible variation. These results are promising because they suggest the possibility of summing diffraction patterns at different angles to reduce acquisition time. In the second part, the characteristics of an EDXRD system based on a pixellated spectroscopic detector and two types of multi-angle scatter collimators (square holes of 1mm side and rectangular holes of dimensions (0. 5mm x 1mm)) are discussed. Both collimators present angles (2°, 4°, 6°, 8°, 10°). Results showed that it was possible to sum diffraction patterns collected simultaneously at different angles for reducing the acquisition time per sample. Although finer collimation allows better momentum transfer resolution, this is not a stringent requirement when analysing biological tissue, featuring intrinsically broader peaks. On the other hand, broader collimation allows significantly better statistics. In the final chapter, a feasibility study of diffraction spectroscopic CT is presented. The system comprises a pixellated detector and fan shaped collimator matching the beam divergence. Different scatter images acquired by tilting the detector off-plane. Finally, the potential of combining the two approaches (simultaneous acquisition of multiple angles and one-shot profile) is discussed. It can be expected that the combination of the two concepts could reduce the acquisition time to ~15min/sample. This would make diffraction CT an appealing instrument to be used for fast tissue characterization in operating theatres and pathology laboratories.