Abstracts

Rationale: Source localization of interictal epileptic discharges (IEDs) is clinically useful in the presurgical workup of epilepsy patients. Recently, our group has demonstrated that distributed magnetic source imaging (MSI) has better accuracy than clinically approved equivalent current dipole method (ECD) (Pellegrino et al., 2018). Here, we aimed to compare the performance of four different distributed source imaging techniques on MSI: Minimum Norm Estimate (MNE), dynamic Statistical Parametric Mapping (dSPM), standardized Low-Resolution Electromagnetic Tomography (sLORETA) and coherent Maximum Entropy on the Mean (cMEM, an entropy-based technique) (Heers at al., 2016; Chowdhury et al., 2016). Methods: Source localizations of IEDs (N=359) derived from 50 focal epilepsy patients who: underwent pre-operative magnetoencephalography recordings, had a well-defined focus determined through intracranial EEG and epileptogenic MRI lesions. Presence and location of an underlying MRI lesion, as well as extension and outcome post-surgery were also analyzed. The comparison of different MSI techniques was based on the minimum distance between the MSI peak and the focus and the assessment of spatial dispersion, a measure of the spatial spread of spurious activity outside the epileptic focus. Results: Spatial dispersion was significantly lower for cMEM (mean±SD in mm:19.44±13.47) as compared to all other methods (dSPM:34.74±8.50; MNE:35.40±10.81; sLORETA:33.85±9.69). MSI accuracy, determined as the minimum distance between the source maximum and the epilepsy focus, was clinically comparable across methods and ranged within few millimeters (dSPM:10.41±12.31; MEM:11.49±17.07; MNE:9.73±17.73; sLORETA:7.93±13.40) (Figure 1). Conclusions: Distributed MSI is an accurate tool for localizing the epileptic focus. cMEM provides the lowest amount of spurious activity, while obtaining similar localization accuracy compared to other techniques. This is an important feature in particular for MRI-negative patients that might have results of MSI identifying targets for invasive EEG investigation, as well as for determination of surgical candidacy.References: Pellegrino G et al., Hum Brain Mapp. 2018 Jan;39:218-231. Heers M et al., Brain Topogr. 2016;29:162-81. Chowdhury RA et al., PLoS One. 2013;8:e55969. Funding: Canadian Institutes of Health Research; Natural Sciences and Engineering Research Council of Canada; Fonds de Recherche du Quebec–Sante; American Epilepsy Society. Dr. Pellegrino is supported by the MNI Frederick Andermann Epilepsy and EEG Fellowship, sponsored by an unrestricted educational grant from UCB Canada.