A Design-of-Experiments Approach to FEM Uncertainty Analysis for Optimizing Magnetic Resonance Imaging RF Coil Design

J. T. Fong[1], N. A. Heckert[1], J. J. Filliben[1], L. Ma[1], K. F. Stupic[2], K. E. Keenan[2], S. E. Russek[2]
[1]National Institute of Standards & Technology, Gaithersburg, MD, USA
[2]National Institute of Standards & Technology, Boulder, CO, USA
Published in 2014

The design of a magnetic resonance imaging (MRI) RF coil using finite element method-based analysis is an essential part of a multi-year research project at the National Institute of Standards and Technology, Gaithersburg, Maryland, where the goal of the project is to develop a “phantom” for calibrating MRI machines that is traceable to standardized values. Traceable MRI calibrations are expected to enable accurate, quantitative measurements of tumors and other disease markers that can be reproduced across many different patients, scanners and clinics over time --- and potentially reduce medical costs. In this paper, we first define, YC, a dimensionless metric for the non-uniformity of the magnetic flux density norm profile, and then used a design-of-experiments (DOE) approach to optimization. The figure below shows the results before and after the application of the DOE approach.