Basics Introduction

1

Dear Users,

I know this is not a real topic for a forum. I am very new to brainstorm, and to some extent to EEG too. I need desperately some basic help. For training, I am trying to process some EEG data (edf/mff). I do not have the anatomy so I need to use a template. I also do not have the coordinate of the electrodes but we use a default cap (EGI GSN 256). Could anyone show me step-by-step how to configure the system to perform preprocessing.
Thanks a lot

Luigi

2

Hi Luigi,

This forum is the perfect place to ask any question about Brainstorm.
If you are planning to use Brainstorm, start by following all the introduction tutorials, at least until #19 (or until #23 if you're planning to do some source-level analysis). It's about MEG, not EEG, but it will teach you how to use the software. Most of it is very similar for EEG.
https://neuroimage.usc.edu/brainstorm/Tutorials

Then read the tutorial EEG/Epilepsy, it gives additional information about EEG:
https://neuroimage.usc.edu/brainstorm/Tutorials/Epilepsy
Then pick in the advanced tutorials anything else you'd need to learn about.

After this week of training, you should be able to start with processing your own recordings.
Let us know if you need additional directions at that moment.

Enjoy the reading!
François

3

Hi François,

Thank you so much for your reply. I feel kind of ashamed to tell that I have already looked into the tutorial but didn't find what I was looking for. Even the topics you referenced in your mail didn't help. At this point I do think I need a step-by-step explanation.

Luigi

4

I didn't mean to "look into" I really meant to "follow" these tutorials: you download the example dataset and you do everything without questioning it too much.
You'll realize all the pre-processing (and much more) is explained step by step. This will be some useful reading for you, especially if you're new to the field.

Then you'll have only three things to changed from what you did in these tutorials:

  • In the configuration of the subject you create in the database: use "Default anatomy: Yes"
  • When you link the continuous recordings to the database: change the file type to "EEG: EGI-Phillips"
  • When you set the position of the electrodes: use one of the available instead of the channel files.

"I need step-by-step explanation for pre-processing EEG" is not a question focused enough to be addressed on a forum.
If you already read everything and could not find what you were you looking for, please tell us exactly what you were looking for, where you looked it up, what is missing precisely, and we might be able to help you further.

Cheers
Francois

5

Thank you Francois. I appreciate. I had red the tutorial specifically the sections concerning my doubts. I could

  1. create a protocol with Yes Yes checked.
  2. create the subject.
  3. I loaded the default anatomy template.
  4. I don't think I have to do the MRI fiducials if I use the predefinited cap, right? Neither I have to do anything else with registration, right?
  5. Right click on subject -> Use default cap -> select the cap. Correct?
  6. Right click on subject -> review raw file to load the recording, Right?

Best regards,

Luigi

6

I don't think I have to do the MRI fiducials if I use the predefinited cap, right?

Correct, use the fiducials already set in the template.

Neither I have to do anything else with registration, right?

Indeed, the default electrodes positions in Brainstorm are already aligned on the head surface of the template anatomy.

Right click on subject -> Use default cap -> select the cap. Correct?
Right click on subject -> review raw file to load the recording, Right?

No, first create your link to the continuous file, then right-click on the channel file > Add EEG positions.
https://neuroimage.usc.edu/brainstorm/Tutorials/Epilepsy#Prepare_the_channel_file

Note that our advise is really to follow all these online tutorials using the datasets provided before trying to process personal datasets. We've been having a lot of positive feedback in the past 10 years with this approach, you would probably end up saving a lot of time on the long run.