Abstracts

Rationale: Epilepsy patients undergoing intracranial EEG provide a unique opportunity to further examine seizure propagation with the combined use of EEG and functional MRI (fMRI). In preparation for investigations involving humans, we assessed the MR safety of implanted intracranial depth electrodes at 3T.Methods: A phantom model was constructed to emulate the shape, size and conductivity of the human head (Figure 1), in which a commercially-available intracranial depth electrode was implanted. Measurements were obtained from the implanted device to examine movement, temperature and induced current through a series of MR scanning conditions. Device movement was measured in a standard head coil using two separate methods to determine both translational and rotational forces. Temperature measurements were obtained from two locations on the depth electrode, a reference location within the phantom, and in the air external to the phantom. Temperature was monitored and recorded consistently over each scanning session using a four-channel, fibre-optic thermometry system. Induced voltages during scanning, and voltage frequencies were measured using a digital oscilloscope and for a variety of phantom and probe positions. All measurements were performed at 3 Tesla using a GE Signa scanner under the following MR scanning conditions: T1-weighted 3-plane localizer, 2D anatomical, T1-weighted 3D anatomical, high-order shimming, fluid attenuation inversion recovery (FLAIR), gradient-recalled echo planar imaging (GRE-EPI) and T2*-weighted GRE-EPI fMRI (explored with TR of 1000ms, 1500ms, and 2000ms). Results: The intracranial depth electrodes experienced no measurable rotational or translational deflections throughout the entire scanning period, except a 2° change due to table movement as it advanced into the scanner bore. Temperature changes from depth electrodes and surrounding phantom tissue were less than 0.5°C throughout all scanning conditions. All induced voltages oscillated at the Larmor frequency of approximately 127MHz, and ranged from 200-2800mV depending on the scanning conditions (induced voltages were greatest for FLAIR and DWI/DTI, then fMRI sequences and lowest for T1-weighted anatomical images and high order shim sequences). Induced voltages using the depth electrode with commercial EEG-fMRI system oscillated at 60kHz or 120kHz with a peak voltage of 450-650mV.Conclusions: Temperature changes and device movement were negligible and well within conservative safety guidelines. The amplitude of induced voltage was within the physiologically active range; however, the frequency is much higher than the <10kHz required to produce neuronal stimulation. These results suggest that intracranial depth electrodes should not pose a risk in performing intracranial EEG-fMRI at 3T.