A topology optimization method based on the solid isotropic material with penalization interpolation scheme is utilized for designing gradient coils for use in magnetic resonance microscopy. Unlike the popular stream function method, the proposed method has design variables that are the distribution of conductive material. A voltage-driven transverse gradient coil is proposed to be used as micro-scale magnetic resonance imaging(MRI) gradient coils, thus avoiding introducing a coil-winding pattern and simplifying the coil configuration. The proposed method avoids post-processing errors that occur when the continuous current density is approximated by discrete wires in the stream function approach. The feasibility and accuracy of the method are verified through designing the z-gradient and y-gradient coils on a cylindrical surface.Numerical design results show that the proposed method can provide a new coil layout in a compact design space.
A general pre-processing procedure of the measured SAXS patterns for reducing the effect of beam stop and beam stop holder is described. A proper method for automatically choosing the regularization parameter is implemented. The method works out on the two-dimensional SAXS patterns. After deblurring, the corresponding two-dimensional patterns will be converted into one-dimensional integrated intensity distribution curves. We tested the program using both calculated artificial data and real experimental data such as polystyrene and poly(methyl methacrylate) latices. The deblurred results are satisfactory showing the effectiveness of the method. The deblurring process of a typical two-dimensional SAXS pattern using the Matlab based program can be completed in few seconds on normal personal computers.
