LAUSR

Introduction Source localization of interictal epileptic discharges (IEDs) is clinically useful in the presurgical workup of epilepsy patients. Recently, we have demonstrated that distributed magnetic source imaging (dMSI) has better accuracy than clinically approved equivalent current dipole method (ECD). Here, we aimed to compare the performance of four different dMSI techniques: 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). Methods We analyzed dMSI results of 206 IEDs derived from MEG recordings in 28 focal epilepsy patients who had a well-defined focus determined through intracranial EEG, epileptogenic MRI lesions or surgical resection. dMSI accuracy and spatial properties were quantitatively estimated as: (a) minimum distance between the source peak and the focus; (b) within-subject reproducibility; (c) spatial dispersion of the source map outside the focus; (d) extension of cortical map; (e) effect of thresholding on map size and properties. Results Distance between the map peak and epilepsy focus as well as within subject reproducibility were clinically comparable across methods (median distance from the focus around 1 cm). Spatial dispersion was significantly lower for cMEM. cMEM maps display typically higher contrast between the source maximum and surrounding regions, being therefore less sensitive to map thresholding. Conclusions All dMSI techniques under investigation provided excellent performance in localizing the epileptic focus. cMEM provides the lowest amount of spurious activity, while obtaining similar localization accuracy compared to other techniques. dMSI techniques currently available for clinical use have the potential to significantly improve identification of intracranial EEG targets and to guide surgical planning.