Answers are grounded in this paper and cite their sources.
The homogeneity of magnetic field is one of the most critical parameter of magnetic resonance image (MRI) magnet. Passive shimming (PS) technique is widely used to correct the inhomogeneity of the basic field. The conventional PS methods are always based on the hypothesis of saturation magnetization and linear programming. However, due to engineering approximation, nonlinear magnetization characteristic of shimming pieces (SPs) and special field condition in 0.7-T biplanar superconducting MRI, the error of conventional method is too large. Hence, a novel PS method based on magnetic coupling (MC) model and niching genetic algorithm (NGA) is presented in this paper. MC model is proposed to calculate the magnetizations in SPs instead of statured-magnetization hypothesis. To solve this complex and large-scale optimization problem, NGA is employed, which can avoid the error caused by rounding and improve the effect of optimization. During the procedure of optimization, the magnetizations of each individual are update in each iteration. The theoretical and experimental results show that the MC model and the NGA are effective and reliable.
No takes yet. Share an insight, caveat, or question.
Zhu et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: