Head-only MRI of Deep Brain Stimulation devices at 0.5 T: Patient-derived approaches for estimating radiofrequency heating

Neuroimage. 2026 Sep 18:122244. doi: 10.1016/j.neuroimage.2026.122244. Online ahead of print.

ABSTRACT

The growing prevalence of neurodegenerative diseases has increased the number of patients treated with deep brain stimulation (DBS), many of whom will require magnetic resonance imaging (MRI) during their care. However, radiofrequency (RF)-induced heating of conductive DBS leads during MRI remains a primary safety concern, often restricting imaging options or denying patients access to MRI altogether. Head-only 0.5 T MRI may offer a safer alternative due to intrinsically lower specific absorption rate (SAR) and reduced coupling efficiency between the RF field and DBS leads compared to clinical field strengths (1.5 and 3 T). This study presents experimental RF heating measurements of two bilateral DBS systems in an ASTM phantom at 0.5 T, alongside a patient-specific heating prediction pipeline that combines benchtop transfer function measurements with electromagnetic simulations (using Sim4Life) of the RF exposure to quantify electric fields along the DBS lead trajectories (extracted from postimplantation CT scans). Phantom testing demonstrated that nearly all sequences produced heating below the 2°C regulatory threshold for both a legacy DBS device (Activa PC with 3387 leads) and a modern MR-conditional device (Percept PC with SenSight™ leads) up to a B1+RMS of 4 μT. Patient-specific predictions, incorporating conservative uncertainty estimates, also remained below this safety threshold. These findings support head-only 0.5 T MRI as a viable and potentially safer option for routine neuroimaging of DBS patients, enabling access to advanced protocols such as diffusion tensor imaging that may be impractical at higher field strengths due to RF heating constraints.

PMID:42759781 | DOI:10.1016/j.neuroimage.2026.122244