I have recently used the Emotiv Xavier Testbench software (emotiv.com) to convert .edf files to .csv - maybe try looking into that. I would then also recommend EEGLAB, since you can obtain toolbox extensions for both mentioned formats, and then proceed to follow Stephen's advice.

https://emotiv.com/store/app/

Finally, how do you solve this problem?

@Wei Li

I had to use the nihon Kohden software to do it.

Sincerely thank you very much. Did you  convert a lot of .eas files to  .edf files by the nihon Kohden software or collect EEG data again by Nihon Kohden?

It was some time ago, but I think you could convert it from there or there is also a nihon plugin for eeg lab. It will require some time.

Stephen Politzer-Ahles

I have trid nihon plugin for EEGLAB, but it failed to open the .eas file. Except for the nihon Kohden software, is there any open source toolbox that import the .eas files. Thanks.

Muhammad Zubair Ahmad Xiaochen Tang

I have the same problem. The CADWELL software only allows the conversion one by one, which is not efficient. I tried the Nihon plugin in EEGLAB and failed as well. Any open source toolbox to read .eas files as Xiaochen asked? That would be a live saver! Thank you!

I'm planning to include support for eas-cadwell files in Biosig http://biosig.sourceforge.net.

In order to do so, I'm looking for example files. These can be uploaded to

https://seafile.ist.ac.at/u/d/89201a8a5dab41d4b51d/

Ideally, these examples should be acompanied by the same data converted into some other (open) file format. That would enable testing the correctness of the Biosig implementation. Alternativly a short description (sampling rate) and screenshot of a certain section might also help.

Alois Schlögl Glad to learn that you are planning to work on this. I am asking my colleague to prepare some .eas sample file and will upload in the next a few days. BTW, do you have a plan to support EZ3/Arc files from Cadwell as well? we have different cadwell EEG systems and they generate different kinds of files (.eas/.rec/.ez3/.Arc) ? I don't understand why they make things so complicated. Thank you very much for your help in advance!

If you provide examples, I'll check what is possible.

Alois Schlögl Thank you very much for your willingness to looking into this! Sorry for my a bit late response. I uploaded the original .eas /.ez3 /.ezdata to your "eas-cadwell-example-files" folder. The eas and ez3 files have been converted to .rec and .edf files respectively. Besides, the sampling rate, durations, and the number of channels are included as well. However, we have trouble converting .ezdata into .edf or .rec file. Nevertheless, I provide one example to see whether you are able to read it (trying my luck). Let me know if you need more information. Thank you again!

Thanks, I received your files. Because I do not have the specifications for these file formats, the format specification need to be reverse engineered. This does not seem impossible, but it would say about 2weeks to 2 month (full time effort) should be sufficient to tell whether a (basic) decoder can be build. Let me know if anyone wants to contribute, or if you are interested in this as well. Moreover, I'd prefer if the discussion about further details could be moved to email (you can find my address in the source code of Biosig and on the web) or the biosig mailing.

If you are interested in such a solution, you can support this effort through the crowd-funding campaign as described here: https://liberapay.com/aschloegl/

Has there been any update on this? I am looking for a way to convert the data from CadWell machine to anything compatible with MATLAB or python