[FieldTrip] Dipole Fitting Algorithm

[Mehmet-Akif Coskun]

Dear Fieldtrippers,


I am having little trouble understanding the dipole fitting algorithm. Any comments will be greatly appreciated.


1) During dipole fitting, a model data (Vmodel) is created using the computed leadfield. How exactly this model data is computed? Is it simply the projection of the pinv'ed leadfield to the true data (Vdata)? 


2) The error is computed between the model data and true data for the best dipole. Is this RMS error?


3) The last part of dipole fitting is a non-linear search. I really wonder what is the algorithm, what exactly non-linear search is doing?


Thanks in advance
Mehmet


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20110531/68e3ea38/attachment.html>