AUTHOR=Fujita Yuya , Khoo Hui Ming , Hirayama Miki , Kawahara Masaaki , Koyama Yoshihiro , Tarewaki Hiroyuki , Arisawa Atsuko , Yanagisawa Takufumi , Tani Naoki , Oshino Satoru , Lemieux Louis , Kishima Haruhiko 

TITLE=Evaluating the Safety of Simultaneous Intracranial Electroencephalography and Functional Magnetic Resonance Imaging Acquisition Using a 3 Tesla Magnetic Resonance Imaging Scanner

JOURNAL=Frontiers in Neuroscience

VOLUME=Volume 16 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/neuroscience/articles/10.3389/fnins.2022.921922

DOI=10.3389/fnins.2022.921922

ISSN=1662-453X

ABSTRACT=Background: The growing interest in understanding ongoing human brain activity and networks in health and disease, and the unsurpassed sensitivity of intracranial electroencephalography (icEEG) make the simultaneous icEEG and functional magnetic resonance imaging (fMRI) acquisition an attractive investigation tool. However, safety remains a crucial consideration, particularly due to the impact of the specific characteristics of icEEG and MRI technologies and their combined configurations on the associated health risks. We assessed the health risks and feasibility of our icEEG-fMRI protocol using a clinical 3-T scanner with body transmit and head-receive coils.

Methods: Using platinum and platinum-iridium grid and depth electrodes implanted in a custom-made acrylic-gel phantom, we assessed safety by focusing on three factors. First, we measured radio frequency (RF)-induced heating of the electrodes during fast spin echo (FSE, as a control) and the three sequences in our icEEG-fMRI protocol. Heating was evaluated with electrodes placed orthogonal or parallel to the static magnetic field. Under the configuration with the highest heating observed, we measured the total heating induced in our protocol, a continuous 70-min icEEG-fMRI session comprising localizer, echo-planar imaging (EPI), and magnetization-prepared rapid gradient-echo sequences. Second, we measured the gradient switching-induced voltage under configurations emulating electrode implantation in the frontal and temporal lobes. Third, we assessed the gradient switching-induced electrode movement by direct visual detection and image analyses.

Results: RF-induced heating in the vicinity of the icEEG electrode contacts tested were on average greater in the orthogonal than parallel configuration, with a maximum increase of 0.2°C during EPI and 1.9°C during FSE. The total local heating was below the 1°C safety limit across all contacts tested during the 70-min icEEG-fMRI session. The induced voltage was within the 100-mV safety limit regardless of the configuration. No gradient switching-induced electrode displacement was observed.

Conclusions: Under the conditions reported in this study, we provide evidence that the additional health risks associated with heating, neuronal stimulation, or device movement are low when acquiring fMRI at 3 T in the presence of intracranial clinical electrodes available in Japan. High specific absorption ratio sequences such as FSE should be avoided to prevent potential inadvertent tissue heating.