collecting emotions based on specific EEG events
elizabeth
los angeles
[original title of thread was: "display each channel separately"]
Dear OpenBCI Lead developer,
@retiutut We looked at the current software and for our project we need to have a details of each channel displayed separately. We need a widget that shows all 16 channels separately. It is very important.
Re: Correspondence
Jul 20, 2020, 10:49 PM EDT
Hi Lisa,
Thanks for your email. RW, our lead GUI developer, can address your questions.
Please post the three questions to the technical forum, in the section called software (or GUI,), and tag RW using @ retiutut . (no space after the @ )_
best,
My initial email.
Sam
What I did not figure out
1. How to display each channel separately? Like if I need to have details of channel 1 in a separate frame?
Do you think I have to use MatLab or other software what can do it? Or I shall figure out and write a custom widget?
1. How do we expand a time frame from 5 seconds to let say one or 10 minutes?
2. How do we change the settings?
I look forward to hearing from you.
Sincerely,
Lisa Smirnova
(310) 279-6967 cell
Comments
1. The next public release of the GUI will allow you to select what TimeSeries channels you would like to view, not just disabling them.
It would be very bad to display 10 minutes of TimeSeries data using Processing and the Graphica library. It would be incredibly slow, even though we have already modified graph library to be faster.
2. Solution: To view insights into what's happening minute-by-minute, try using the Spectrogram Widget! This widget shows FFT over time.
You are welcome to make recordings using the GUI, and load the CSV files into Matlab. You can even access boards directly in Matlab without the GUI by using the BrainFlow Matlab binding, found here: https://brainflow.readthedocs.io/en/stable/BuildBrainFlow.html#matlab https://brainflow.readthedocs.io/en/stable/Examples.html#matlab
Take Care,
RW
Thank you very much. Could you please, let me know how soon a new public release will be. This is for my work project and is very important.
Elizabeth, you can always view your recording in EDFBrowser, which is a general purpose EEG viewer with many features. With no limitations on recording length. EDFBrowser can parse CSV files, but can also read 'BDF' files, which the GUI can output.
https://www.teuniz.net/edfbrowser/
Thank you very much. If a release to be done soon with each channel separately, this is all I need for now. - I am left with a questions what software best will fit my needs and where cvs file records and if openbci software records it because as I looked I saw only text files with numbers of data. - I am looking learning how to use it since Friday. So it means I still work to collect all knowledge puzzle together.
Somehow I did install the software and run data stream, but seem not to understand how to connect EDFBrowser. It does not see it.
I checked the repository. Could you please approximately tell me the time frame for the next public release. It's very important to me for my work. I just will try to continue to work overnight to download all computable software to seeing if I can look separately at each channel. I will also look for practical applications and differences of all related software on the market because to me its not obvious of the benefits of each separate. But eventually I figure it out.
Give me a moment and I will make a pre-release.
Thank you!
https://github.com/OpenBCI/OpenBCI_GUI/releases/tag/v5.0.0-beta.2
Now, there is a "Channel >" button in the top left of the TimeSeries widget. Clicking this opens a small dropdown menu where you can select which channels you would like to have visible.
Take Care,
RW
Thank you. Will be testing now. It works, and shows each channel separately. Could you please, send me exe file because the processing is programming tool and it is a bit complicated to run it in front of my boss. You make me become a programmer. which I don't mind.
I just wonder if it is possible to create an executable file. I care for business people to use it because I had to download java environment and follow instructions to run processing. It could be much better just having an executable file. Thank you!
@elizabeth, hi.
I merged your NEW thread, into this existing thread on the same subject. Please try to continue conversations on existing threads, rather than opening many new threads. This will speed responses from previous comment participants. As they get email notifications.
You can also private message any user by clicking on the username, then the message button.
William
https://github.com/OpenBCI/OpenBCI_GUI/releases/download/v5.0.0-beta.2/openbcigui_v5.0.0-beta.2_windows64.zip
Thank you.
1. iS the FFT based in the time periods 20,10, 5,3,1 seconds. Its the FTT of the 200 (10,5,3,1) second time period.
Can we expand time of recording to 20 minutes, one hour, two hours. 8 hours.
What if we have an experiment where we record brain waves for 2 hours and export it to the excel.
Can we select an event and get FFT data of the event? Can we have window with time is FTT Period
I included screenshots
Thank you!
Elizabeth, hi.
The OpenBCI_GUI unfortunately cannot be a general purpose program that has all features possibly needed by researchers. Generally OpenBCI users that need features such as you discuss, use other apps or conversion utilities that are more in alignment with their needs.
https://docs.openbci.com/docs/06Software/SoftwareLanding
Given what you have written above, you may want to investigate OpenViBE, EEGLAB (Matlab), BioEra. In terms of a general purpose viewer program that can window in throughout a long recording, see EDFBrowser. (GUI can output BDF, which can be read by EDFBrowser.)
https://www.teuniz.net/edfbrowser/
Regards, William
The GUI 'Networking Widget', is capable of streaming (to another app), up to three different data streams from the GUI.
https://docs.openbci.com/docs/06Software/01-OpenBCISoftware/GUIWidgets#networking
So for example one stream could be your EEG time series, and another stream could be the FFT data. More FFT details in some threads such as,
https://openbci.com/forum/index.php?p=/discussion/2426/phase-information-from-fft-plot-networking
Thank you for the comments and links, William.
I am now trying to understand how to make the scale from 20 to 30,000 work. Its my requirements I was given at work, and I am not quite sure what it measures. My understanding that uV are my (or whomever uses the equipment) electric waves. I looked at what is measured by alpha, betta, etc and the range is different. Could you give me any links to the websites to read what is being measured.
Could you just send me links to any programs that can help you measure in these ranges. The rest is clear and I will have a MATLAB soon. I downloaded OpenVibe, and its a lot of information.
My goal is to make sure I come up with the solution that lay audience can use the program in the easiest way possible. Thanks again.
Sincerely,
Lisa
I just go google what is 30,000 uV and it does not give me explanation what is being measured.
What is the goal of your application? What are you trying to measure? What was "my requirements I was given at work"?
EEG may seem like a precise science, but it is more akin to holding a microphone up to the crowd noise at a football stadium. You are picking up the summed activity of billions of neurons. And each electrode picks up from a wide area, not just directly under the electrode. More like several square cm.
Also clarify this statement, where are you getting the "30,000" figure. Generally EEG uV are in the range of several uV to say 80 uV max. Anything greater than that can be considered noise from artifacts.
Do please note that Cyton CSV file DOES NOT have any filters applied. So there will be a high DC offset, generally in the range of some millivolts, ~10 millivolts (~10,000 uV) is not uncommon. Your signal processing will generally remove this by doing either a highpass at .5 Hz or a bandpass, say from .5 Hz to 45 Hz. Additionally a notch at your mains frequency (50 or 60 Hz) is helpful.
https://openbci.com/forum/index.php?p=/discussion/201/large-millivolt-data-values-fbeeg-full-band-eeg
My boss tells me those requirements per each completed phase. Generally it is a neuroscience project and I just need to prepare the program for the next phase of requirements. the 30,000 is the last uV that my requirements are. Maybe the project requires measuring the noise form artifacts. That is why I am stuck - I need to explain why these numbers cannot be on the scale and find a solution.
I need to do all kind of research before asking question why. I just need to learn if it is even possible and what will be the highest and the lowest range possible.
PLEASE tell me the requirements and what you are trying to achieve, see previous questions I posted.
To me, now it is clear what can be done with other programs and I did experiment with OpenVibe and I will download a free trial of MATLAB before asking for the license.
https://www.google.com/search?q=eeg+digital+signal+processing+tutorial
Matlab may be overkill, particularly if you have to purchase a commercial vs student license.
PLEASE tell me the requirements and what you are trying to achieve, see previous questions I posted.
I listed all requirements above.
I will not pay for a MATLAB because in case I will be given to me.
MATLAB seems to be a main stream program and best to analyses data.
WHAT REQUIREMENTS ARE:
1. I need to show PPT Plot of selected time frame events, which should be possible in MATLAB.
2. I need 2-8 hours recordings,
3. I need a range of uV form 20 to 30,000 on the scale.
4. I probably need to save a data stream to play it back.
5. I need to teach lay audience how to use the program.
Obviously my plan is to learn in my free time all the programs and see what will best fit.
2-8 hour recording for one session.
Thank you for helping me.