On the estimation of the worst-case implant-induced RF-heating in multi-channel MRI
- 16 May 2017
- journal article
- Published by IOP Publishing in Physics in Medicine & Biology
- Vol. 62 (12) , 4711-4727
Abstract
The increasing use of multiple radiofrequency (RF) transmit channels in magnetic resonance imaging (MRI) systems makes it necessary to rigorously assess the risk of RF-induced heating. This risk is especially aggravated with inclusions of medical implants within the body. The worst-case RF-heating scenario is achieved when the local tissue deposition in the at-risk region (generally in the vicinity of the implant electrodes) reaches its maximum value while MRI exposure is compliant with predefined general specific absorption rate (SAR) limits or power requirements. This work first reviews the common approach to estimate the worst-case RF-induced heating in multi-channel MRI environment, based on the maximization of the ratio of two Hermitian forms by solving a generalized eigenvalue problem. It is then shown that the common approach is not rigorous and may lead to an underestimation of the worst-case RF-heating scenario when there is a large number of RF transmit channels and there exist multiple SAR or power constraints to be satisfied. Finally, this work derives a rigorous SAR-based formulation to estimate a preferable worst-case scenario, which is solved by casting a semidefinite programming relaxation of this original non-convex problem, whose solution closely approximates the true worst-case including all SAR constraints. Numerical results for 2, 4, 8, 16, and 32 RF channels in a 3T-MRI volume coil for a patient with a deep-brain stimulator under a head imaging exposure are provided as illustrative examples.Keywords
Funding Information
- Universidad Autónoma de Madrid (2015/ASIA/03 (UAM-Banco Santander Programme))
- Ministerio de Educación, Cultura y Deporte (JC2015-00116 (José Castillejo Programme))
- Comisión Interministerial de Ciencia y Tecnología (TEC2013-47106-C3-2-R)
This publication has 22 references indexed in Scilit:
- Convex optimization of MRI exposure for mitigation of RF-heating from active medical implantsPhysics in Medicine & Biology, 2015
- Development of a new generation of high-resolution anatomical models for medical device evaluation: the Virtual Population 3.0Physics in Medicine & Biology, 2014
- Parallel transmit pulse design for patients with deep brain stimulation implantsMagnetic Resonance in Medicine, 2014
- Evaluation of the RF heating of a generic deep brain stimulator exposed in 1.5 T magnetic resonance scannersBioelectromagnetics, 2012
- A specific absorption rate prediction concept for parallel transmission MRMagnetic Resonance in Medicine, 2012
- Local SAR management by RF Shimming: a simulation study with multiple human body modelsMagnetic Resonance Materials in Physics, Biology and Medicine, 2011
- Semidefinite Relaxation of Quadratic Optimization ProblemsIEEE Signal Processing Magazine, 2010
- Graph Implementations for Nonsmooth Convex Programs2007
- Convex Optimization2004
- The Approximation of One Matrix by Another of Lower RankPsychometrika, 1936