The EEG/MEG reconstruction of brain activity in cortical areas close to the skull has made enormous advances in the last decades. However, the recovery of brain networks involving deep-lying sources is still a challenging task for any inverse method. Depth mis-localization, a well known error of many methods, and the separation of single sources in multiple-source scenarios are very important issues for clinical applications, e.g., for presurgical epilepsy diagnosis. Hierarchical Bayesian modeling (HBM) emerged as a unifying framework for current density reconstruction (CDR) comprising most established methods as well as offering promising new methods. Our work examines the performance of HBM for source configurations consisting of few, focal, potentially deep-lying sources when used with realistic, high resolution Finite Element (FEM) head models and EEG sensors.

@InProceedings{LPBW11,
  author       = {Lucka, F. and Pursiainen, S. and Burger, M. and Wolters, C.},
  title        = {Hierarchical Bayesian Models for EEG Inversion: Depth Localization and Source Separation for Focal Sources in Realistic FE Head Models},
  booktitle    = {Biomedical Engineering},
  volume       = {56},
  year         = {2011},
  publisher    = {De Gruyter},
  issn         = {0939-4990},
  url          = \{/2011/LPBW11},
}