Meeting Banner
Abstract #2722

An Automated Framework to Decouple PTx Arrays with Many Channels

Zohaib Mahmood1, Bastien Gurin2, Elfar Adalsteinsson1, 3, Lawrence L. Wald, 23, Luca Daniel1

1Dept of Electrical Engineering & Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States; 2Martinos Center for Biomedical Imaging, Dept. of Radiology, Massachusetts General Hospital, Charlestown , MA, United States; 3Harvard-MIT Division of Health Sciences Technology, Cambridge, MA, United States

In this paper we present an algorithm to automatically design a decoupling matrix for pTx arrays with many channels. The algorithm optimally selects the decoupling matrix based on the constraints defined on network such as reciprocity, passivity and the network being lossless. We show that our algorithm converges and the decoupling matrix achieves ideal decoupling. We explore various useful features of the decoupling network by analyzing L-curves and singular values. However the problem of realizing the optimal decoupling matrix using practical circuits is still open and is currently work in progress.

Keywords

achieve adjusting allows amplifiers array arrays audience automated automatically biomedical block body build capacitance channel channels circuit coefficient coil coils components compute computed condition conditions consider constraints consumption convergence converges cost coupled coupling curves decoupled decoupling degenerate depicting dept design designed destructive development distributed electrical enforcing ensure error every excitation expressed fidelity fields forward framework function future general greatly guesses health ideal ideally impedance implementing inductance initial input least limiting linear loaded local loss lossless lost lumped magnitude many matched matching mathematically matrices matrix mixed mixes model nearly network networks norm note objective ones optimal optimally optimized original paper partially passive patient patterns perfect performance placed power problem produces proposed pulse pulses random realized reciprocal reciprocity reflected require required residuals response robustness rows secondary selects sensitivity shimming side simple simulation singular solution solutions solve special specification square structure support symmetry system target technology thirty tissue topologies types uncoupled uniqueness unlike unreliable