Importing data from ANT to Brainstorm

Discussions
1

Hi Francois,

I am having trouble importing EEG ANT data.

  1. It does not recognize the channel file in the raw file (I have to import it manually from the default ANT channel file, and I do not know which one is the right one).
  2. It does not import the events from the .evt file.

Could you please help me with that?

Thank you

Simone

2

Hi Simone,

The file reader we use in Brainstorm comes from the FieldTrip program (libeep version 3.3.167):

I can see that there was a new release of the libeep library (3.3.173):
https://sourceforge.net/projects/libeep/
Maybe using this library instead would help reading some newer file formats, but I don't think it would help with your current problems.

  1. The .cnt files do not include the position of the electrodes. If you want to add this information in the Brainstorm database, it is something you have to do manually.
    Either you digitize the position of the electrodes (example: http://neuroimage.usc.edu/brainstorm/Tutorials/TutDigitize), or you use a default set of positions.
    If you are not sure which cap to use from the Brainstorm interface, or if you don't think Brainstorm provides correct positions for your EEG cap, please contact ANT and ask them if they can share a default set of electrodes positions. I will include it in the distribution.

  2. About the events:
    So far, all the EEProbe example files I have include an event file with the extension .trg, not .evt.
    Where are these .evt files coming from?

Cheers,
Francois

3

Hi Simone,

I could try to add the support for this file format in Brainstorm, but I would need a full example dataset: the matching .cnt and .evt file.
Could you send me this? (upload the files somewhere, then send me the link in separate email - click on my username on this forum)

Thanks,
Francois

With respect to the event file, I attach here the default file (.evt)
that is generated along with the EEProbe (.cnt) by the ANT system.

I also exported the raw EEG file recorded with ANT system as a
Neuroscan .cnt file. In this way, I've been able to read it in EEGLAB
along with the correct marker positions (but not in Brainstorm).
Finally, I've been able to correctly import the .set and .fdt files
generated by EEGLAB in Brainstorm.

Is there any way you could implement the .evt format in Brainstorm?

4

Hi Francois,

I’m following up on this old post to see if you were able to implement an import process for the .evt files generated by the ANT system? Their older software (asalab) used to generated a .trg file, but their new software (eego) does not. I believe the tiggers are embedded in either the .cnt file or the .evt file. Without them, I’m not able to epoch the data. Any suggestions for a work around would be great.

Thanks for your help!
Ann

5

Hi Ann,

Sorry for the response delay.

No, I don’t think this has ever been addressed, I don’t have any example ANT .evt files.
The .cnt binary files have the capability to store the events. Do you know why your events are not embedded in the .cnt?

I guess the format of the .evt files is not simply the BESA file format, but it’s worth the test. Have you tried adding the events to the recordings with the file filter “BESA (*.evt)”? (tab Record > menu File > Add events from file).

If the .evt are text files or if you can find Matlab code to read them, I could add their support in Brainstorm.
If you would like me to have a look at it, please send me an example dataset (matching .cnt and .evt files). Upload it somewhere and send me the link in a separate message.

Francois

6

Hi Francois,

Thanks for your reply! I have tried exporting the data in multiple different file formats, but none of them work quite right. The triggers are supposed to be embedded in the eeprobe.cnt file, but I think they are stored in an unused channel that might get stripped out of the header when it is exported (I think but am not sure). I also tried reading .evt file with the BESA file format, but that doesn’t work either. I’ve contacted ANT but have not found a solution yet. I’d be happy to share sample files with you to see if you can help with a work around.

Thanks again!
Ann

7

Hi Ann,

Thank you for the files.
Brainstorm was indeed not using correctly the information coming from the .evt files.
I think I fixed the issue. Please update Brainstorm (menu Update > Update Brainstorm), and try again importing your file in Brainstorm.

Cheers,
Francois

8

This is great - thank you so much for the quick response! It does import the events, but they are not organized by segment (i.e., condition). The recording software allows us to run several conditions in sequence without stopping and starting. The same same event labels (i.e., 1, 2, 3, 4) are used for each segment, so they need to be sorted by segment - if that makes sense. I think there is a “segment” file that might need to be integrated with the import? Or maybe that can be done through the Import Events command? Let me know if you have questions.

Thanks much!
Ann

9

I don’t know how to interpret this .seg file…
This information is not read by the EEGLAB plugin I use for reading the file.

In the file, it says “NumberSegments=7”, but there are only 6 entries. And what are the values below? Is the 3rd value the duration of the segment in samples?

If you give information on what the values in the .seg file mean, I could possibly read the file as multiple segments, but it will not make it much easier to handle in Brainstorm. Maybe naming the events differently only could help (“1-2” for segment 1 event 2 for instance).

Another option is that you epoch them separately in Brainstorm: first import segment 1 by entering in the time window field the timing of segment 1, it would update the list of events and display only the ones in this segment. After importing, rename them. It makes the process very manual…

10

We agree - we don’t know how to interpret the .seg file either. It looks like the event code “0” may contain the time stamp for both an impedance check AND the new segment. If so, the new segment would be at every odd numbered event currently coded under the “0” labeled event. I’m also providing a link to another file that is generated when we export using the BrainVision file type. This may have additional information regarding how the events are being coded in the .evt and/or .cnt files. Hopefully this will help!

http://aeddins.myweb.usf.edu/files/MA_2017-11-17_13-55-55.txt

11

Hi Ann

I use ANT Neuro as well, and had this issue with importing events.

The way the support people helped me to get around the issue was as follows;

  1. Export the EEG data as a .cnt file from Eego.
  2. Export the EEG data again as a Brainvision file (.eeg) from Eego. This creates several new files, including a .vmrk file.
  3. Import the EEG data from the .cnt file in Brainstorm.
  4. Once you’ve got the link to the raw file open, select File -> Add events from file from the events panel in the Record tab, and select the .vmrk file. This will load all of the events from all of the segments recorded.

Luli

12

If you can export to the BrainVision format, it would be simpler to process everything using only these files (.eeg+.vhdr+.vmrk).

Ann: I’m sorry, I haven’t time to look at the files you sent me yet. It may take a few weeks before I explore this issue further. In the meantime, maybe you could try this export to BrainVision/BrainAmp.

13

Hi Luli -

Thanks so much for the suggestions! We have been able to import the events using the .vmrk file, but we have to organize the events into segments manually. A little added work, but doable.

Thanks,
Ann

14

Hi Francois - No worries about the delay! We are continuing to try and solve it with the BrainVision export option.

Thanks again,
Ann

15

Hi Ann,

I finally looked at the files you sent me but could not understand the correspondence between the “New segment” markers in the BrainVision .vmrk files and the values in the ANT .seg files.

If at some point you can get this information (maybe asking to the ANT customer support), I could read the .seg file when creating the link in Brainstorm, and add a marker “New segment” as it is done for the BrainVision export (which still requires a manual selection of the import time window), or I could split the file and make the various segments available as different epochs in the file viewer (not every easy to manipulate in Brainstorm).

Francois