I have processed EEG data with EEGLAB and ERP data with ERPLAB. Now I would like to estimate the source of the following ERPs: P1, N170/VPP, EPN, and LPP. After a lot of reading I've narrowed it down to source estimation techniques available in Cartool and Brainstorm. But I can't make up my mind. Which software do you recommend for ERP source estimation of these two? Where can I find a good tutorial? Brainstorm provides a nice tutorial, but couldn't find one for Cartool.
Thanks in advance!
I'm not really familiar with Brainstorm, nor with Cartool, for source localization I use Fieldtrip toolbox. It's a Matlab-based toolbox, and it has very good documentation and hands-on tutorials as well.
http://www.fieldtriptoolbox.org/tutorial/beamformer/
I'm not really familiar with Brainstorm, nor with Cartool, for source localization I use Fieldtrip toolbox. It's a Matlab-based toolbox, and it has very good documentation and hands-on tutorials as well.
http://www.fieldtriptoolbox.org/tutorial/beamformer/
My lab has used multiple processing tools and we have settled on Brainstorm as the optimal answer. I notice that Fieldtrip has also been recommended, and I agree that it is a good choice. But if you are not a strong matlab programmer it may be more difficult to get that up and running in the lab. Brainstorm also has good and useful tutorials
I think your question can be generalized for all source modelling software packages and for magnetic signal sources also. Whatever the signal source is, electric potential distribution or magnetic field distribution, all attempts to find a best fitting source model are suffering from the "inverse problem". This means there are infinite possible solutions for source compositions which can explain the experimental data in a more or less well fitting way. One of the most important suppositions is to start the modelling with a single dipole, multi dipole or distributed kind of dipoles approach. Comparisons of data sets from different individuals can help a lot to find the best explainable one. But the final goal should always be to find a model which can best explain your measured data by means of minimizing the difference between measured sensor data and predicted data of your chosen model (and point in time ore epoch). Different software solutions will be more or less "handy" during processing of data but in my opinion they all should provide a way of easily calculating a measurement telling you this difference for a point in time or within a time window.
I agree with Linda. Brainstorm is probably the most user friendly and has lots of available support / resources / tutorials to help you learn how to use it.
- Badges
- Science topic
- Similar topics
- Psychology
- Cognitive Science