DIPFIT plugin for dipole source analysis with EEGLAB
DIPFIT is a plugin for the EEGLAB Matlab toolbox. Using the plugin, you can fit dipoles to the topographical potential distribution of independent components. The plugin is described in more detail on http://www.sccn.ucsd.edu/eeglab/plugins.html.
Balazs Vegso 14:35 on November 29, 2005 Permalink
Dear Robert!
I’m using components of your dipfit plugin (lead field calculation) for my research. Of course your work will be cited in the article i’m preparing. I would like to know if there is any way of me receiving the equations used in your leadfield calculation? I’m referring to eeg_leadfield*.m which states that the equation is taken from Luetkenhoener,1992, which is not available online on the web (for now i’m not yet a registered IEEE member, so i cannot download the publication possibly containing this equation). So I was wondering if you could send me the mentioned equations for the 1- and 4-layered spherical models.
Sincerely,
Balázs Végső
PhD student
University of Veszprém
Hungary
Robert 15:07 on November 29, 2005 Permalink
Dear Balázs
The DIPFIT plugin consists of several components, each of which ahs been published elsewhere. One part of the implementation consists of the forward model, which can be single- or four-sphere models for the EEG, the boundary element method for the EEG, single sphere for MEG or multisphere for the MEG. Other aspects are the non-linear optimization method (for which I am using the Matlab optimization toolbox) and, most importantly, the general concept of using a dipole model for solving the EEG or MEG inverse problem
The one-sphere and four-sphere forward model are originally described in “R. Kavanagh, T. M. Darccey, D. Lehmann, and D. H. Fender. Evaluation of methods for three-dimensional localization of electric sources in the human brain. IEEE Trans Biomed Eng, 25:421-429, 1978″. The description there is only given for a dipole along the z-axis and electrodes at arbitrary locations. To compute the solution for a dipole at an arbitrary location, I am using a coordinate transformation that transforms the dipole and all electrodes simultaneously, so that the dipole ends up at the z-axis. This coordinate transformation is described in “Lütkenhöner, Möglichkeiten und Grenzen der neuromagnetischen Quellenanalyse, 1992″. The coordinate transformation is clearly not as crucial as the solution of the forward problem itself, therefore you only have to refer to Kavanaugh. The (german) book by Lütkenhöner was printed in small quantities and is not available through regular channels, which is another reason why it is not so usefull to refer to.
The general concept of using nonlinear optimization with a dipole model to explain the EEG potential distribution is originally best captured in the bookchapter “Scherg, M. (1990). Fundamentals of dipole source potential analysis. In: Auditory evoked magnetic fields and electric potentials (eds. F. Grandori, M. Hoke and G.L. Romani). Advances in Audiology, Vol. 6. Karger, Basel, pp 40-69″.
best regards,
Robert