Use this URL to cite or link to this record in EThOS:
Title: Multiple particle tracking using the Birmingham positron emission camera
Author: Gundogdu, Ozcan
ISNI:       0000 0004 2714 8670
Awarding Body: University of Birmingham
Current Institution: University of Birmingham
Date of Award: 1999
Availability of Full Text:
Access from EThOS:
Birmingham Positron Imaging Centre has the only known Positron Camera dedicated solely for industrial processes. It has been successfully applied to numerous industrial processes as a non-invasive imaging method such as oil flow in engines, flow in rocks and extrusion processes. A novel technique of particle tracking has been developed at the University of Birmingham using a single radio nuclide labelled particle as a tracer. This technique allowed non-invasive monitoring of industrial processes such as mixing, gas fluidised beds and rotating drums. It provided important information about the effects that govern physical behaviour of the particle inside an industrial machine during its natural working cycle. This information can be used in designing and building more efficient, full scale industrial plants. However, if more than one particle could be traced simultaneously, more information such as interaction between the particles could be gathered and this could be used to identify the physical processes that are taking place. This study is about tracking multiple particles-particularly two particles. A number of methods such as positron emission holography, cluster analysis and tomography have been considered and related computer codes have been developed. The principals of Positron Emission Tomography (PET) and Positron Emission Particle Tracking Method (PEPT) have been explained in Chapter 2. The PEPT algorithm is discussed with some of the applications. Chapter 3 gives an overview of zone plate coded imaging technique that makes it possible to image incoherent radiation that would not be possible with reflective and refractive methods. It reviews some of the important zone plate designs and makes use of an ideal zone plate. Furthermore, a description of a reconstruction computer code that simulates diffraction is given. Chapter 4 gives the application of the zone plate encoded holography, Positron Emission Holography (PEH) technique, to the PET trajectories obtained experimentally for two particle tracking for the first time. The limitations of this technique have been evaluated for a different number of trajectories and resolution of the system was investigated. The trajectories obtained should ideally cross at the location of the particle. Since there are two particles, it is sometimes possible to group these trajectories into two groups of particles. The techniques of grouping are mainly found in the literature under the title of cluster analysis. It is for this reason, that Chapters gives a review of the principles and comparison of the different clustering techniques available in the literature. Chapter 6 describes a unique way of representing the trajectories. It is the midpoint of the closest distance between the trajectories. It also describes a clustering method that will cluster these midpoints into two clusters, rather than trajectories themselves,in order to locate the two particles. The advantages of this algorithm in comparison to the techniques given in Chapter 5 are illustrated with examples. Some of the difficult questions such as how well the objects are clustered or how many natural clusters can be formed by these objects that arise in cluster analysis are also investigated. The application of the clustering algorithm described in the above chapter is given for two stationary positron particles in Chapter 7. The application is repeated for different particle distances, for different particle orientations and a different number of trajectories. Chapter 8 describes another clustering method known as Multiple Location-Allocation method that is mainly used in economics or resource management. This technique is applied to the voxel reconstruction of the trajectories. A computer code for this technique was developed and wide range of techniques were considered from Digital Image Processing for thresholding. The application of the above technique to the experimental data for two stationary particles is given in Chapter 9. The initial allocation techniques were tested and a simple but very effective way of thresholding was described. In Chapter l0, all the above techniques were employed in tracking moving particles. One of the particles was placed on a rotating table which can be rotated at a constant angular speed. The other particle was fixed to a stationary position. The PEH algorithm does need a high number of trajectories but this puts limitations on the accuracy of the particle locations since during the time it takes to collect that number of trajectories, the particle might have moved to another position. However, the clustering algorithm developed in Chapter 6 provided very accurate results making it possible to track the particles for very short time periods. The algorithm developed in Chapter 8 also provides very accurate results but the PAM algorithm can work with a smaller number of trajectories. Finally Chapter l0 gives a summary of the conclusions and makes suggestions for possible future work.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available