A Design Tool for Magnetic Resonance Imaging Gradient Coils Using DUCAS with Weighted Nodes and Initial Current Potentials

作者：Abe Mitsushi^*

来源：IEEE Transactions on Magnetics, 2013, 49(12): 5645-5655.

DOI：10.1109/TMAG.2013.2275411

• Current potential
• gradient field coil
• MRI
• optimization
• singular value decomposition

DUCAS is a computer code using an arbitrary magnetic field distribution as input and producing a sheet current distribution as output. The code was improved for application to the magnetic resonance imaging (MRI) gradient coil (GC) design. DUCAS computes the current potential (CP) distribution, and the coil conductors are placed along the contours, which are current flow lines. An improved algorithm with weights on node CPs and input initial CPs is described for use with DUCAS. The algorithm obtains the CP distribution by summing singular value decomposition (SVD) eigendistributions with the initial CP distribution. The weights modify the eigendistributions. The initial CP distribution is used to get smooth flow lines and then coil patterns. The improved DUCAS is suitable for avoiding a complex current distribution and to obtain coil patterns that are easy to manufacture. The initial CP distribution is assumed first, then DUCAS compensates the CPs for the given magnetic field distribution with the weights. This paper describes the computational algorithm and shows a GC design obtained with it.

• 出版日期2013-12