Abstracts

RATIONALE: Patients who have undergone brain surgery challenge the head models used in dipole modeling of EEG interictal spikes. Indeed, they present brain resections and skull defects, which influence the propagation of electric fields. METHODS: We used Boundary Element Model (BEM) realistic head models that include a skull open burr hole, a hole filled with acrylic or a brain resection. For each model, we simulated dipoles in the vicinity of defects, and measured the impact of not including the defect on dipole reconstruction. RESULTS: For the open burr holes, the localization error was up to 10mm and the amplitude was overestimated by a factor of up to 2. For the burr holes filled with acrylic, the localization error was similar and the amplitude was underestimated by a factor of up to 2. For the resection the effects were much smaller, with mislocalization of 2mm and negligible amplitude error. CONCLUSIONS: The fact of not including post-surgical burr holes can lead to significant errors in dipole modeling. The boundary element method may be used in a clinical setting to model these skull defects and minimize these errors. CSF cavities, whether surgical or pathological, do not seem to have a significant influence.