Abstracts

Rationale: Ultra-high field (≥ 7 Tesla [T]) magnetic resonance imaging (MRI) has potential important implications for assisting with intracranial investigations of epilepsy such as stereo-electroencephalography (SEEG) due to increased contrast and signal compared with standard field strengths. We sought to investigate the feasibility of integrating 7T imaging into the pre-surgical workflow to assist with accurate SEEG targeting in patients with refractory epilepsy. We aimed to achieve this goal in two stages: first by evaluating the impact of 7T geometric distortion on the ability to perform accurate targeting, and second by integrating a high resolution (600 micron) normative 7T template into the SEEG planning workflow. Methods: In 22 control subjects scanned at both 7T and lower field, we evaluated local geometric changes using deformation-based morphometry (DBM), which permits voxel-wise discovery of sub-millimeter to millimeter level distortions. Differences between 7T and lower field were evaluated using nonparametric Wilcoxon rank sum testing. We previously created a 7T template for assisting with deep brain stimulation target selection, and have extended our deformable registration workflow for template-to-patient fusion in a series of patients investigated with intracranial electrodes. Electrode contacts within the hippocampus were semi-automatically labeled from post-operative computed tomography (CT) scans, registered with the pre-operative planning MRI, further propagated into 7T template space, and finally transformed into a recently developed intrinsic hippocampal subfield coordinate system for group evaluation of ideal contact location. All p-values were corrected for multiple comparisons correction using the false-discovery rate (corrected p-value < 0.025). Results: Results from morphometric analysis are summarized in Table 1. Minimal distortion was observed in the bilateral hippocampi (effect size: 0.72 +/- 0.20 mm; not significant). However, pertinent regions such as the bilateral temporal poles, inferior temporal gyri, and fusiform gyri showed statistically significant distortion (effect size: 1.2-1.6 mm; corrected p-value < 0.025) (Table 1). Bilateral inferior mesial frontal lobes and bilateral subcallosal cortices also demonstrated distortion (1.7-1.8 mm; corrected p-value < 0.025). From a total of 101 SEEG cases performed at our centre, we performed 7T template-to-patient registration in a cohort of 51 subjects with first time SEEG implantation including the hippocampus proper. This deformable registration process enabled the evaluation of ideal electrode contact locations relative to hippocampal subfields in 7T template space and a novel unfolded hippocampal coordinate system proposed by our group. Conclusions: Ultra-high field MRI enables an unprecedented level of anatomical detail achievable in in vivo subjects. Susceptibility artifacts and distortions remain problematic, but a multi-scale imaging approach using both standard and ultra-high field magnets may prove optimal for pre-operative planning for epilepsy investigation. Funding: The presenting author is supported by the Western University Clinical Investigator Program accredited by the Royal College of Physicians and Surgeons of Canada and a Canadian Institutes of Health Research (CIHR) Frederick Banting and Charles Best Canada Graduate Scholarship Doctoral (CGS-D) award. This project was supported by a CIHR Project Grant (Funding reference number: 148839), EpLink - The Epilepsy Research Program of the Ontario Brain Institute, and scanning was performed at Western’s Centre for Functional and Metabolic Mapping, supported by the Canada First Research Excellence Fund to BrainsCAN.