NeuroImage (Nov 2024)

Feasibility of using toroidal transceivers for acquiring intraoperative MR images around deep brain stimulation electrodes

  • Nur Izzati Huda Zulkarnain,
  • Alireza Sadeghi-Tarakameh,
  • Russell L. Lagore,
  • Dee M. Koski,
  • Gregory J. Metzger,
  • Zuzan Cayci,
  • Noam Harel,
  • Yigitcan Eryaman

Journal volume & issue
Vol. 302
p. 120912

Abstract

Read online

Introduction: Magnetic resonance imaging (MRI) provides excellent soft tissue contrast for visualizing of deep brain stimulation (DBS) targets, allowing validation of the electrode placement, and assessing complications such as microhemorrhage and edema. However, the presence of the electrodes can introduce challenges such as radiofrequency (RF) induced current artifacts and excessive heating of the electrode contacts. Additionally, extended procedure times are also considered a disadvantage when using MRI as an intraoperative imaging modality following DBS electrode placement. Method: We propose a novel approach of using toroidal resonators to inductively couple the shaft of the electrode to the scanner's transmit-receive chain thereby utilizing it as a localized imaging antenna. The small extent of the field generated by the electrode antenna allows fast imaging with smaller field-of-views (FOVs) spanning only a few centimeters. Furthermore, we present a fast and accurate safety monitoring strategy that can be used to predict the temperature increase at the electrical contacts of the electrode. Results and Discussion: Imaging with the toroidal transceiver yields a higher signal-to-noise ratio (SNR) efficiency in proximity to the electrodes. This approach reduced the RF induced current artifacts around the electrode which enhanced the visibility of the shaft and improved electrode localization. Moreover, the limited sensitivity around the electrode can be exploited to perform fast scans with small FOVs. The predicted heating around DBS contacts was in quantitative agreement with the experimental heating in swine studies with a normalized root-mean-square error (NRMSE) ≤ 0.09.

Keywords