Investigation of excitation coils configuration and particle swarm optimization of coil’s diameter in magnetic nanoparticles imaging

• Main Author: Ng, Kent Hoo
• Format: Thesis
• Language: English
• Published: 2015
• Subjects: R Medicine (General)
• Online Access: http://eprints.utm.my/id/eprint/53965/19/NgKentHooMFBME2015.pdf; http://eprints.utm.my/id/eprint/53965/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:86110

Magnetic Nanoparticles Imaging (MNI) is a new imaging method which uses magnetic nanoparticles as the contrast agent. The superparamagnetism of ferro- and ferrimaterial allowed the particles to act as a temporary magnet when an external excitation is applied. The imaging is based on the reconstruction of the measured relaxation time (magnetorelaxometry) from the sample by SQUID sensors after the applied field is switched off. To date, the imaging method is still not applicable on human due to many undetermined parameters of the diagnostic configuration. The equation of the reconstruction is generally ill-posed due to the involvement of many parameters in the configuration. Hence, the optimization is required in order to eliminate the unstable parameters of this imaging technique. In this thesis, the study of excitation coils’ configuration has been conducted with manipulation of the number coils, arrangement of coils, diameter of coils and the shape of coils. The results show that by increasing the number of coils will improve the correlation coefficient of the reconstruction. However, due to the limitation of the grid size, the size of the coils is another concerned criterion; hence, the total area of the coils in each grid must be smaller than the area of the grid. The correlation coefficients of the reconstruction with varying the coils diameters show an improvement with smaller coils diameter. This observation obeyed the algorithm of diameter optimization in this thesis. Besides, the examinations of the arrangement and shape of coils show that both these criteria are highly depending on the shape of the sample. Due to the size of the rectangular coils used in the simulation are basically larger than the circular coils, hence, better results are found in the circular coils. On the other hand, the examination of the possibility of applying particle swarm optimization (PSO) method on the coils diameter has been studied by constructing the algorithm of the magnetic field calculation. The results of the algorithm construction showed that the diameter element is a non-linear parameter, and it is not suitable to be used in PSO as this method only applicable for the element which poses a linear relationship between the element and the output of optimization.