Hello Francois,
I need to report a bug regarding segmentation. When I tried this with the Run>Import function, each epoch seems to be fixed to 1 second. Then, I segmented the continuous file by right-clicking on "Link to raw file", this worked fine. I tried with other files, and the results were the same. Can you take a look into this?
Sungjin
I tested this with a continuous EEGLAB .set file, with the same options as in your screen capture.
In output, I get correctly epochs of [400-2000]s.
Are you sure you are not trying to import a file that is already epoched?
I you can't figure out what the problem is, please send us a small (and zipped) example file with which you are having issues (upload it somewhere and post the download link here)
Hello Francois,
This is a continuous file, not epoched. As you can see in the screenshot, I had no problem epoching this file when I tried this manually. I hope you can figure out why, and once again, thanks for your help.
https://www.dropbox.com/s/gkdnhlr6bnusj21/0303_Cleaned_v4.set?dl=0
This error has happened to me when another processed file already existed within the file tree.
I recommend checking to see if you don't have any other files in the tree with a 0-1s time definition.
If you do, try either moving them, or deleting them.
Cheers from a fellow user,
Jared
Thanks for the file, I could understand what your problem is.
The events you defined in this file are extended events with a 1s duration. In this case, the option "Epoch time" is ignored, and all the event is imported.
If you want to import longer epochs, you need to either create simple events instead (with no explicit duration), or convert the existing extended events to simple events with the process "Events > Convert to simple events" (it can be called directly from the Record tab, in the Events menu).
Hello Jared,
Thanks for your advice. I didn't delete files in the file trees because I haven't finished analyzing existing files. Once I am finished, I will definitely try your method.
Sungjin
Hello Francois,
I am confused because I am sure that I converted all extended events to simple ones when I first created existing epochs manually. When I checked individual files, none of them are extended events. Plus, the file duration is over 30 minutes. (see the screenshot).
Can you look into the possibility that this is a bug as Jared suggests above?
Sungjin
What is available in your EEGLAB file are 1s-long events. If you try to import them directly into Brainstorm, it would create 1s epochs, regardless of the "Epoch time" option you set in the input parameters.
To be able to import longer epochs, you need to:
My guess is that you are trying to import your epochs from the .set file instead of the modified "Link to raw file" available in the Brainstorm database. This is what the first screen capture your posted show: if you want to import epochs from files already available in the database, the option "Files to import" in the process options should be empty (not set to the EEGLAB .set file...)
Hello Francois,
Thanks for your support and patience with me. As you suggest, I tried to import epochs from files already available in the database by not selecting the original Set file in the "Files to import" box. However, this gave me an error message "No file selected". You probably understand what I am trying to achieve here, and I wonder if you can provide a sample script?
Sungjin
You need to select the files you want to process in Process1 before calling the process:
https://neuroimage.usc.edu/brainstorm/Tutorials/PipelineEditor
Hello Francois,
After multiple correspondence between us, I was finally able to epoch imported data files. Thanks for your help and have a good weekend!
Sungjin
