# Question

**URL:** <https://mne.discourse.group/t/question/1672>\
**Category:** Mailing List Archive (read-only)\
**Tags:** list-archive\
**Created:** [December 5, 2018, 4:01pm UTC](https://mne.discourse.group/t/question/1672 "2018-12-05T16:01:40Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/85cc6bd2b69cb698a166dc6d880fb550510d0144.jpeg) [@system](https://mne.discourse.group/u/system)\
**Post date:** [December 5, 2018, 4:01pm UTC](https://mne.discourse.group/t/question/1672/1 "2018-12-05T16:01:40Z")

</div>

External Email - Use Caution

Hello MNE!

I am working on EEG based Brain Computer Interface and using MNE as  
analysis tool. The EEG recording device which I am using saves the  
recordings in eeg format. I convert it to edf format before loading to  
python as MNE doesn't support eeg format. I want to make a real time  
BCI using python language. Is there any way such that I can acquire  
raw EEG signal in python and real time BCI developed using MNE? Please  
help me regarding this as I feel completely lost.

Regards  
Manoj Kovvuru  
M.Tech in Robotics  
IIEST Shibpur

---

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/85cc6bd2b69cb698a166dc6d880fb550510d0144.jpeg) [@system](https://mne.discourse.group/u/system)\
**Post date:** [December 5, 2018, 4:25pm UTC](https://mne.discourse.group/t/question/1672/2 "2018-12-05T16:25:08Z")

</div>

External Email - Use Caution

Hello!

MNE does support .eeg files created with a BrainProducts amplifier (mne.io.read\_raw\_brainvision). However, if you have a different amplifier, converting to .edf seems like a good idea.

Regarding developing a real-time BCI, MNE only supports offline analysis. You might want to check out [https://github.com/sccn/labstreaminglayer](https://github.com/sccn/labstreaminglayer), which you can use to stream data across a network (there are Python bindings available). You could then build a signal processing pipeline using custom Python code (you could of course use scikit-learn for machine learning and scipy for signal processing, I think they support online applications).

Clemens

---

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/85cc6bd2b69cb698a166dc6d880fb550510d0144.jpeg) [@system](https://mne.discourse.group/u/system)\
**Post date:** [December 5, 2018, 11:09pm UTC](https://mne.discourse.group/t/question/1672/3 "2018-12-05T23:09:57Z")

</div>

External Email - Use Caution

MNE does support real-time analysis, although it is not so well advertised  
and tested for EEG systems.

You can check out a few examples here  
\<[https://martinos.org/mne/dev/auto\_examples/index.html#real-time-m-eeg-acquisition](https://martinos.org/mne/dev/auto_examples/index.html#real-time-m-eeg-acquisition)\>.  
It's essentially a wrapper around the FieldTrip buffer  
\<[http://www.fieldtriptoolbox.org/getting\_started/realtime/](http://www.fieldtriptoolbox.org/getting_started/realtime/)\>.

Best,  
Mainak

---

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/85cc6bd2b69cb698a166dc6d880fb550510d0144.jpeg) [@system](https://mne.discourse.group/u/system)\
**Post date:** [December 6, 2018, 6:58am UTC](https://mne.discourse.group/t/question/1672/4 "2018-12-06T06:58:58Z")

</div>

External Email - Use Caution

Interesting, I didn't know about that :-).

Clemens

---

<div class="post-metadata">

**Author:** ![system](https://global.discourse-cdn.com/free1/uploads/mne/original/1X/85cc6bd2b69cb698a166dc6d880fb550510d0144.jpeg) [@system](https://mne.discourse.group/u/system)\
**Post date:** [December 6, 2018, 12:58pm UTC](https://mne.discourse.group/t/question/1672/5 "2018-12-06T12:58:35Z")

</div>

External Email - Use Caution

Hi Manoj,

If I understand correctly what you are trying to do, you should check out  
mne-cpp:

[https://www.mne-cpp.org/](https://www.mne-cpp.org/)

Here's an example SSVEP BCI application:

[http://wiki.mne-cpp.org/index.php/Setup\_SSVEP\_BCI\_example](http://wiki.mne-cpp.org/index.php/Setup_SSVEP_BCI_example)

what kind of hardware are you working with?

Cheers,

-Morgan

PS I am seeing now that the MNE examples using the fieldtrip buffer are  
somehow connected/overlapping projects. I'd love to learn more about both.  
I hope the mne-cpp links help though.