Intracranial-electroencephalography (icEEG) is a specific invasive EEG recording technique with the aim of precisely identifying focal epileptogenic networks in patients with drug-resistant epilepsy. In recent years, there has been an increased interest in monitoring single unit activity during epileptic seizure using microwire electrodes protruding at the end of intracranial depth electrodes. Performing high-resolution MRI post implantation could significantly improve the accuracy of anatomical electrode localisation, with the risk however of radiofrequency (RF) induced heating in the vicinity of the electrodes. Although a number of studies have used this approach without report of adverse incidents ^1-7, systematic safety assessment has been limited. Our purpose was therefore to assess temperature changes (∆T) in the vicinity of microwire electrodes (AdTech, Racine, USA) with reference to our established safety protocol for standard icEGG ⁸ on our 1.5T MR system.A torso phantom was filled with a gel of poly-acrylic acid partial sodium salt, as described previously5. Temperature recordings were made using a fibre optic temperature thermometer (Neoptix, Canada). Temperature probes were attached to the tips of 2 depth electrodes positioned either parallel or perpendicular to B₀. Additional probes were positioned 5 mm from the distal end of the depth electrodes. The tails of both electrodes were arranged such that they were parallel to B₀ and isolated from each other. Measurements were performed at 1.5T (Avanto, Siemens, Erlangen, Germany) using a transmit-receive head coil, and temperature increase (∆T) was calculated during a 6 min Turbo Spin Echo sequence (scanner-estimated head SAR: 3 W/Kg, B1_RMS 4.3µT). Microwire electrodes were then removed and measurements were repeated at the same probe locations.Maximum ∆T was measured for the macro/micro electrode parallel to B₀ and representative results are shown in figure 1: Maximum ∆T rose to +0.5°C at 5 mm from the distal end of the depth electrode, in the area where microwires are, however ∆T remained below our experimental thermometry precision (≤0.1 °C) when the bundle of microwires was removed. ∆T rose to 0.4°C at the tip of the depth electrode parallel to B₀.We observed a small increase of temperature of the gel due to the presence of microwires. This contrasts with a report by Hefft et al.⁵, who with their arrangement did not measure any temperature increase on a 1.5T scanner using a body-transmit/ 12 channel receive head coil during a number of sequences (maximum 3.9W/Kg scanner reported SAR). Heating did not exceeded 1°C in any of our measurements with our specific set up, supporting the view that under controlled conditions MRI with these electrodes in situ may be safe. However caution may still be warranted, as due to the microwire size (40 μm diameter) our measurements of gel heating over a relatively large volume in the vicinity of a bundle of microwires may not fully reflect local heating over smaller volumes at the tip of the microwire itself.