OpenViBE forum

The OpenViBE community
It is currently Sun Apr 30, 2017 12:49 pm

All times are UTC




Post new topic Reply to topic  [ 4 posts ] 
Author Message
PostPosted: Sat Apr 08, 2017 9:24 pm 
Offline

Joined: Tue Jan 19, 2016 12:11 pm
Posts: 3
hello,
I need to use openVibe for the following purposes:
1. harvesting EEG data with Mitsar-202 before neurofeedback training.
2. calculating individual upper alpha (IAF) value after cleaning the harvested data (i.e. the value of alpha frequency at its peak amplitude during recording + 2Hz) - as to research recommendations, this value has to be calculated due to the variance that have been found in alpha values between subjects. Usually, I do this manually (with WinEEG software) but I found it so inconvenient since it takes so long to analyze the EEG data and to remove artifacts while the subject waiting.
3. measuring the upper alpha power.
4. Neurofeedback for IAF enhancement for 8 blocks ; 3 min each with 1 min break between the training blocks.

I built a simple scenario for acquiring the EEG signal (without calculating IAF) but I am not completely sure its a good one neither training blocks.
May you please help ? :shock: :?

thank you in advance!
Nehai


Top
 Profile  
Reply with quote  
PostPosted: Mon Apr 10, 2017 7:43 am 
Offline

Joined: Tue Dec 04, 2012 3:53 pm
Posts: 506
Location: INRIA Rennes, FRANCE
Hello Nehai,

that would need some work as we don't have classic neurofeedback training example. From what I can see, you need

- a component/technique to clean the data. There's eyeblink removal box ('EOG Denoising') and some filtering boxes provided with openvibe, but we don't have pre-made examples of how they can be used for data cleaning.
- to catch the upper alpha. this you can perhaps do with band pass filter, simple dsp, epoch, average (take band + square + aggregate)
- something to pull the max from the alpha. This can be done with 'percentile' in the 'univariate statistics' box. It computes one value per input buffer (epoch).
- a timeline to control the neurofeedback. in openvibe, this is usually done with lua scripts (e.g. motor imagery and ssvep scenarios)
- a gauge display to show the neurofeedback result. The closest box to that is the Graz Visualization which could be pruned to become a simple gauge. Needs coding.


Good luck,
Jussi


Top
 Profile  
Reply with quote  
PostPosted: Tue Apr 11, 2017 12:31 pm 
Offline

Joined: Tue Jan 19, 2016 12:11 pm
Posts: 3
Dear Jussi,
I thank you deeply for your answer. since I am a new user I have lots of question:) besides my neurofeedback protocol is complicated therefore I need to be sure I am using openVibe the right way...

I think I understand how my Neurofeedback design should be...

1. measuring the alpha baseline values above the six Neurofeedback channels (P3,Pz,P4,O3,Oz,O4) - using the aquisition client box; identitity, reference channel (in Mitsar I use A1 and A2 as reference), channel selector (P3,Pz,P4,O3,Oz,O4), temporal filter (for upper alpha - 10-12; I can change the range manually every exp.).
Do I need multible signal display boxes for each electrode I choose?

2. cleaning data - EOG, EOG diagnoses boxes

3. upper alpha - which I will do it by signal display box.

4.pulling the max alpha - as you suggested I will be doing that by 'univariate statistics' box- if I set the percentile value on 10% for example it means the 10% of my signal values are lower 10th percentile ... it can be used as a threshold, right ?

5. here I need a function that will check if the baseline upper alpha values were larger than the value measured in 4 (i.e. the max upper alpha displayed by the 'univariate box' -
[b]my question is - can I apply this by using three boxes: simple DSP.....univariate statictics ---- DSP - as shown in threshold example at one of the posts? if yes, would it be possible to draw an equation to detect upcoming signal in relation to percentile?


"a timeline to control the neurofeedback. in openvibe, this is usually done with lua scripts (e.g. motor imagery and ssvep scenarios)"
6. are you saying the controlling the time of neurofeedback training can be done by the lua scripts? I am not sure which of the boxes I need here... in other words, how can I design 8 blocks of neurofeedback - 3 min of receiving feedback (which has to be an audio one / music) and 1 min resting (stopping the music feedback)
[/b]
May you please explain how this can be done?
thanks again for your help,


Top
 Profile  
Reply with quote  
PostPosted: Wed Apr 12, 2017 8:11 am 
Offline

Joined: Tue Dec 04, 2012 3:53 pm
Posts: 506
Location: INRIA Rennes, FRANCE
Here's some thoughts,

1. Signal Display by default shows all channels that are given to it. You can limit this by channel selector. If it makes a reasonable neurofeedback display is another question entirely (thats why I was speaking of a gauge). In principle you can insert signal displays anywhere in the signal processing chain to see how the data gets transformed.

4-5. I think the univariate statistics return the value of the computation, for example the percetile. This value would need to be inserted to simple dsp to scale the signal, but I think there's no neat way to do this automatically. One option is to EBML stream spy or CSV Writer to get the value out and then insert it to the simple dsp formula by hand. Maybe there's a more neat way to achieve the same, but not sure.

6. Basically in openvibe experiments and such are controlled by timelines of stimulations (event markers): a stimulation at some point in time is interpreted by the boxes in the scenarios to do different things. See the mentioned bci example scenarios to see how such timelines can be constructed with .lua. Then you need these stimulations to change what your scenario does, e.g. by using them to route the signal or have boxes in general react to them, for example start or stop audio playback.

Usually cooking something bigger in openvibe is achieved by trying to do it, and eventually you manage to find out how it can be done. In the end it is possible that some features are simply missing and need to be implemented or hacked, some way or another.


Happy hacking,
Jussi


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 4 posts ] 

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group
Americanized by Maƫl Soucaze.