Neurofeedback, OpenBCI Boards, and general getting started Questions & Thoughts

MitchMMitchM Appleton, WI, USA
Preface: I am new to the field, I am a programmer by trade and an autistic. The following are questions and thoughts by someone from that mindset. I am also hear to learn and give back with what skills and exp I have since most places where i can go to get training as a neuroscience or psychologist either wont work with an autistic or don't offer any support that is of use and assume I have piles and piles of money lying around. I am barely able to afford this hardware but it seems like the best investment for the cost.

Hey all, 

I picked up the Cyton Board recently with the WIFI Board (eventually plan to get the Daisy module to extend it to 16 channels) and am looking to eventually use it to do neurofeedback training. However searching around on the web and this forum, leads to lots of questions and things that are confusing to someone from the outside of psychology &/or Neuroscience. I figured id ask my questions here so others can benefit from the answers. 

Can you do Neurofeedback with any of the OpenBCI Boards? Why or Why not?

IT seems that there is a common comment being repeated: "You need a minimum of 19 channels to do neurofeedback" ... why? 
I did nerufeedback with Gib Koula, MS, LCSW BCIA-EEG (who is now retired) and only ever used it 2 times, the initial intake, and then the scan done after about 50+ sessions. All other times I had AT MOST, 4 channels on me at any one given time, (which could be done on the Cyton without the Daisy module). Doing a search on this forum yields some interesting posts, I also see something related to this in this thread but that just leads to my next question.

Why do you need expensive brain databases to do Neurofeedback?
my first thought on this is because your doing QEEG (Quantitative EEG) and therefor its the only way to get the info you need to train the brain/do nerofeedback... that said, while it makes sense to look at what the entire brain is doing not just one section, it seems like a rather bazaar approach since everyone's brain is different (but there are obvious similarities amusing there is no physical damage) and saying it must be a minimum of 19 channels to do neurofeedback training, its seems to be similar to approaching maintaining a mile x mile garden by doing it all at once whenever you work on it (yes you need to know what the entire garden looks like and how the crops interact with each other) rather then dividing the garden into 7 segments and focusing on a segment a day. I am guessing that you don't do all 19 channels at once, but if that is the case, then why is it required outside of getting a good map of the brain to see what all regions are doing?  Do we know so little about how the brain interacts that the only way to fix a problem is to make all brains attempt to confirm to an average? What if someone like my self wants to do it (which I have) how does coralling my brain to a standardized mapping help me (spoiler: it kind of did, but it wasn't the practitioners fault.. it was done back in 2008 things have changed since then.) To assume that my Autistic, Dyslexic, ADHD effected brain should be made the same, seems ignorant and counter intuitive... but then I don't have an understanding of the field. I get the impression this isnt the best way... but its the *Safest* way to do this.  

Aside from going to a four year university for neuroscience or psychology (I have tried but its hard to get started when your on the spectrum) what resources like http://learn.neurotechedu.com/,  & https://github.com/NeurotechBerkeley/bci-course would people recommend?  just googling or wikipida-ing it is dangerous when you are new and i love reading but besides these few locations, there is not much i can find that is good for referencing or helping understand why my brain puts out delta waves when I'm awake playing a video game and its known to us outside the field to be used in sleep (see example screenshot of readings from my Cyton while playing a video game I played as a teenager)

all that said, I am just leaning and asking questions with my thoughts unfiltered (sorry for any negativity that may come across, its confusion from lack of understanding and people to ask and basically seeing that there is no financially viable way to help my self that is not less then 100$/hr or the cost of a small house in equipment costs and subscriptions). Thanks also to those who respond. you are also welcome to ask questions of me as I'm just posing my own questions and attempting to move foreword.


«1

Comments

  • wjcroftwjcroft Mount Shasta, CA
    edited July 2018
    Mitch, hi.

    I'll try to answer some of the points in your post:

    ** can you do neurofeedback with OpenBCI boards?

    Yes, BioEra is the most fully featured neurofeedback system currently available for OpenBCI. We are hoping that Bioexplorer will eventually be available, but there have been longer delays than we were originally hoping for. There are other posts on the forum discussing OpenBCI and Bioexplorer issues. Another option is BrainBay. Use the search box or button in the right column to find a tutorial using BrainBay.

    Bioexplorer has the best support for home brain training. This is used in association with Pete Van Deusen's Brain-Trainer web site. You can go there to read more about it. Until Bioexplorer for OpenBCI is released however, that option is not available, I'm very sorry to say.

    ** 19 channels needed?

    No, not at all. Many forms of neurofeedback do not require 19 channels, nor QEEGs. Pete's system can work well with just 2 or 4 channels.

    ** QEEG databases

    These are only sold to licensed clinicians and are very expensive (thousands of dollars, and sometimes annual subscription costs). So not practical for home training. QEEG neurofeedback is only one type, and may not even be the predominant type in the field.  As you mention, there is also the reservation that training the brain to conform to norms from a population standardized database -- may have side effects of dampening one's own unique 'positive' characteristics. Though, if QEEG neurofeedback is done carefully and properly, this is not usually an issue.

    ** brain maps / assessment

    There are other ways of doing assessments than the 19 channel QEEG technique. Pete's system is one of them. Swingle's ClinicalQ is another. Richard Soutar's MiniQ is another. All of these systems are 'safe', and can work with less than 19 channels.

    ** learning

    Pete's system is a great starting point, that is what I would suggest. The other links you mention might not be neurofeedback specific.

    ** delta waves when you are awake

    This is perfectly normal and if you look at the FFT's in the tutorials, substantial delta component is always visible. This of course is where a complete assessment would be beneficial. Which would detect if there is much more delta present "than normal" with either eyes open or eyes closed.

    ** timeframe for Bioexplorer on Cyton / other options

    More details on the Bioexplorer situation on this thread,


    I have no idea if or when Larry Janow will release the beta version he already has in house. He is not communicating with us at the moment. Nor does he communicate with Pete.

    Here is a wild idea if you want to try some kind of workaround, and want to get into the firmware. It's possible to have Cyton emulate OpenEEG. That is a format that Bioexplorer already understands and is public domain.


    It's only 6 channels, not 8. And very small samples, 10 bits, not 24 like OpenBCI. AND, worst of all, it has sample rate of 256 hz, not 250 hz like Cyton. So the level of effort required to do this would be significant. The sample rate being the most problematic. You 'could' just do some trick like inserting 6 extra samples per second. But that would introduce some error in the statistics / signal processing done on that data. 6/256 = ~ 2%, so maybe not that bad. A better way to go would be to 'oversample' at say 500 or 1000 hz, then downsample to 256 (in the firmware). With that approach I believe the PIC32 has enough horsepower to do this but not certain.

    Chip Audette did an (very old by now) version of this OpenEEG emulation for the V1 firmware. I'm not sure he addressed the sample rate issue. Unfortunately it would be unwise to try installing this on Cyton's that have the newer V3.x firmware. So would have to be retrofit. Also, there is the issue of being able to easily switch back and forth between firmware emulations. I don't believe you would want to ONLY do OpenEEG. That would be incompatible with the OpenBCI GUI and all our other mating software. So perhaps on Cyton bootup, you would check a configuration jumper or switch to determine if you are in the optional OpenEEG mode, or the default OpenBCI mode.

    OK, enough for now. 
  • wjcroftwjcroft Mount Shasta, CA
    I just cut/pasted the OpenEEG comments into the other thread about OpenBCI and Bioexplorer. If you want to comment on OpenEEG options, do it there. Since that thread is more devoted to Bioexplorer.

  • MitchMMitchM Appleton, WI, USA
    William,

    Wow! thanks for all this info, I hope others find it helpful also! It is starting to help me cut though the confusion of all this info and get to the core truth of things. (i still do plan to get the Dasiy module as it will give me a better idea of whats in the brain, but wish i could get more mapping points out of the board for a better idea of what is going on all over the head.)  your responses opened up tonds of new questions that flooded into my head.

    so one of the things I am doing is trying to do nerofeedback with LED's hooked into fiber strands. (Example of fiber optics i have on the ceiling, though this configuration is just a first test to see if it will even work/be visible in my apartment.) Image link to V1 fiber base.  but I am switching to a LED grid from Adifriuit Image of grid what I am not sure of since it is Neopixles connected to a particle photon (eventually i am looking at their new mesh network boards which have WIFI, Bluetooth, and Mesh network communications Link to Preorders I am looking to use the Argon boardsbut the BORON LTE lets you do cell data. I also am not sure who on the OPENBCI end that i could work with to get this Arduino board working with my OPENBCI devices, i think it has enough cpu and digital pins (18) that i can work with and a good amount of ram and CPU speed. but since the boards come ALREADY with the Arduino mega baked in, this kind of @#$@#% up my plans.  Yes I plan to opensource it, and if we get this working and approval of OPEN BCI put a guide on Hackster.io for others to follow. Here is a guide I made on there.  Links to other hardware i am looking to use: Adafruit grid i am starting with, Adafruit grid i am looking to upgrade to

    the questions I have on this project are
    ~ Is using only color (potentially sound) a valid way to do neurofeedback?. my gut says yes but i have never seen just this format done.
    ~ What are the different types of neurofeedback?
    ~ If not using the software mentioned above, how would one approach doing the coding for this (i.e. what neurofeedback method would work and how would this training process look like?) my thoughts were something like power of a band (say Delta) red is far away from where the band power needs to be, and Purple, Blue or Green is where the target amplitude(?) should be.

    Other questions:
    This is perfectly normal and if you look at the FFT's in the tutorials, substantial delta component is always visible. 
    Which tutorial are you referring to? what would you recommend for comparing against when researching things on my own as far as "expected wave frequency"?

    Does anyone happen to know how to read charts that show Chemical Shifts (ppm) in the brain. im trying to understand these to charts of my brain better
    1. Active, doing a simple task (push button when image matches) - Link to Image
    2. Eyes Closed, not sleeping but resting - Link to Image

    I know there is an important info in this chart about my autism, but since i cant read it or contact the Dr that made the map, I do not understand it. Also the study this was part of is not yet published ( i have been told they are hoping for end of the year) but i will link to it for others.


  • wjcroftwjcroft Mount Shasta, CA
    Mitch, hi.

    ** feedback modalities

    Biofeedback / neurofeedback can be presented to the user in many different ways. Typically using auditory or visual feedback mechanisms. One of the most common presentations is to show a video movie, and at the same time, control the brightness of the video with the magnitude of the reward feedback that is being given. Auditory feedback is frequently employed with eyes closed sessions, such as meditation training, alpha/theta, etc.

    Please try the BrainBay tutorial to see an example of how this is done with sound. Since everyone has a video screen available on their computers, video feedback or audio feedback are likely much easier to do than the custom hardware you mentioned.

    Generally the simpler the feedback, the better. It's basically an on/off condition you are trying to convey. Different colors and so forth are not really necessary.

    ** different types of neurofeedback

    Look through some of Pete Van Deusen's material. I think you can read the background material on his site. This thread has a number of books you can checkout.


    ** relative band powers

    You can find this in many texts on neurofeedback. Here is one paper on autism training,


    ** ppm charts.

    Sorry, I'm not familiar with this. I don't think it has any connection to EEG or neurofeedback.

    ** QEEG eyes open / eyes closed / Z-score charts

    There are various QEEG tutorials out on the web. 


    The coloring shows how many standard deviations (Z-scores), your map differs from the database standard.
  • MitchMMitchM Appleton, WI, USA
    William,

    Thanks again for your responses. at the very least its helping me parse though things.. 

    *** Feedback modalities
    Yes I was a part of that type of feedback when i did my neurofeedback work. the reason I am working with lights is because I only have sound, and led's available to me in the final environment and a video playing is not yet possible. my thought about colors was to set different reward thresholds using the different colors rather then "on off" just like the brighter and dimmer video screen i had when i was doing my neurofeedback.  but before all this i am trying to take things one step at a time.
     1. Get wifi shield working and sending data.
     2. Lean about parsing the raw eeg data and getting it into something that shows the different bands. (see below)


    *** Band powers & QEEG eyes open / eyes closed / Z-score charts
    Interesting paper, (started to read it a little bit before typing this) this is close to what i was looking for but as i read all the links i find on things, they don't really seem to answer the "normal expected waves and amplitudes in a given region. I'm starting to get some idea of what my head looks like... but most of the stuff in the google search is fluff and selling QEEG not really explaining this. Yes there is going to be a deviation from one brain to another and its impossible to know everyone's brain, but i find it interesting that there is nothing out there to reference a 10-20 site without the expensive brain databases. (even manually parsing it out) and searching for https://www.google.com/search?rlz=1C1CHBF_enUS767US767&ei=Vsk_W7CGEoLNjgTtjZco&q=10-20+eeg+O1+brain+waves&oq=10-20+eeg+O1+brain+waves does not yield much either

    *** Parsing OpenBCI_GUI saved sessions/live data
    this is another one that is perplexing me a bit... i started looking at http://docs.openbci.com/Hardware/03-Cyton_Data_Format which was starting to help me but then I seem to be missing something.  
    when i look at the raw eeg data that the GUI saves from each session i get.. something like

    1, 39076.73, 29028.68, -3653.59, -16416.98, -14594.91, -30709.31, 19061.99, -14459.32, 0.000, 0.000, 0.000, 13:25:41.160

    i see this lovely formula for converting the readings if i am using my own scale Scale Factor (Volts/count) = 4.5 Volts / gain / (2^23 - 1); as well as using the Accelerometer data when using something other then the default  Accelerometer Scale Factor = 0.002 / 2^4; as well as info about parsing the live data as it comes across the stream (24 and 16 bit signed integers) but nothing about how I get the different brainwaves from this so i can duplicate in my own code. the samples i am finding are not for the OpenBCI boards and are missing the starting info for comparison above (Link to example found given parsing info for muse in python)

    Can you help me understand what I am missing here?


  • wjcroftwjcroft Mount Shasta, CA
    Try the BrainBay tutorial. It will explain a lot about detecting different bands. Also, BrainBay uses just the plain Cyton with dongle. So you don't need the Wifi running to use it. 

  • MitchMMitchM Appleton, WI, USA
    yes,

    definatly has the 10-20 to 'general what region does' map i was looking for.

    however the info im looking for converting that number to the band waves is obscured by the BrainBay tools
    so while

    are both great exercises to get started, its lacking the info that i need to do the conversion process in my code. also the tool crashes on my windows 10 laptop even in compatibility mode, so i am stuck trying to figure out what is going on there. 

    I need the wifi shield to work as the Bluetooth drivers the board uses and the examples out there are not well documented so i am looking to the wifi shield to make the process less painful on me. since the arduino is apparently baked into the boards (thus me just wanting the shell and the logic i need to work with the electrodes, means I cannot also use it with my own devices that i know how to work with and am suck using what you guys provide (if this is not the case, i cant seem to see where the site and docs mention this fact...)
  • wjcroftwjcroft Mount Shasta, CA
    Have you looked at Alex Castillo's code?


    I suggest trying BrainBay because it familiarizes you with many aspects of neurofeedback, WITHOUT requiring detailed coding knowledge.

  • MitchMMitchM Appleton, WI, USA
    I appreciate the recommendation and if i can get the program to work, i want to play with it :D

    No I have not yet, I will take a look though his site/blog as there is a lot here. I was hoping this info would be in the OpenBCI Docs but at least you are open to helping me and others with all this stuff. i think https://github.com/neurosity/eeg-pipes might be what I'm looking for but its going to take a bit of piecing together as stuff is spread out though different functions and objects (which what its intent is, was a good choice) but takes a bit to follow to get the actual answer I am looking for... are those raw numbers i have the volts of what the electrodes pick up, or what does that number represent that i need to parse through?
  • wjcroftwjcroft Mount Shasta, CA
    The dongle serial port interface is described clearly on the docs site,


    Cyton is returning the microvolts per channel. There is a DC offset that must be removed, see the BrainBay tutorial and,


    Alex's code is mostly javascript, assuming a connection via wifi.

  • wjcroftwjcroft Mount Shasta, CA
    I'm not sure what glitch you are seeing with BrainBay. Try an earlier release,


    The last, latest release adds support for Ganglion, which you don't need. It also added Hub support, which might be complicating your situation. The previous releases just accessed Cyton via the serial COM port. I think Cyton is still accessed via COM, but Chris might have added an option to access thru the Hub.
  • MitchMMitchM Appleton, WI, USA
    wait so I do not have to do a Fourier transform to the raw data?.... 

    i think i'm confusing my self now so basically...

    1. Cyton gets signal from electrode

    2. stuff happens <-- not 100% sure on this part the coding for Cyton lost me a bit

    3. each channels reading in Microvolts  which is the  1, 39076.73, 29028.68, -3653.59, -16416.98, -14594.91, -30709.31, 19061.99, -14459.32, 0.000, 0.000, 0.000, 13:25:41.160 info that i am looking at now and is what gets sent across the wire.

    4. then taking the microvolts for each channel, i have to _____ to it to convert that microvolts reading into what each wavelength is in Hz(?) or voltage(?) per channel?
  • wjcroftwjcroft Mount Shasta, CA
    PLEASE go through the BrainBay tutorial to understand the signal processing it does. You can also do a web search on DSP digital signal processing.

  • MitchMMitchM Appleton, WI, USA
    edited July 2018
    here is what is happening. i tried installing the BrainBay installer from

    https://github.com/ChrisVeigl/BrainBay/releases

    i originally tired the latest v2.1 but i also juts tried 1.9.3.  it installs just fine, i run the installed program. i see a window, then nothing, (but a process in task manager, after about 2 min, the app listed in task manager just dissiapears.

    so i AM trying to do what you say, BUT when i cant actually follow though with it, it makes it harder to do what you ask. If i can get this program working i am more then happy (and willing) to walk thought the tutorials and demos you sugested.
  • wjcroftwjcroft Mount Shasta, CA
    "window, then nothing...". The window does not have a menu? Sorry I have not tried this recently on Windows 10. If you have a menu, then open the design file given in the tutorial.

  • MitchMMitchM Appleton, WI, USA
    sorry let me reword..

    the app opens up, I have an icon in the taskbar (bottom of the screen) but i cant see anything from it... I thought it might be lagging when trying to load or slow load (not sure why but i have seen weirder things happen) I also open up task manager to make sure i can see that it is in fact running and using CPU (its not, which means its not running or hanging) then the app icon on the task bar and the app listing in the task manager just vanishes, usually indicative of a crash. 

    I tried running it in "compatibility mode" (same area you go to make .exe files always run as admin, which I tried as well)


  • wjcroftwjcroft Mount Shasta, CA
    Could something like this be going on? I have no idea why. Do you have multiple display screens?


  • MitchMMitchM Appleton, WI, USA
    Yes actually I do. its since my main device is a high dp display. i can read things like the OpenBCI GUI. Seems it does not like any external monitors period as having it only use the external or primary (even unplugging the cable) still causes the same issue... tried the arrow keys as well... I think i foresee a format + reinstall in my future... 

    the part of me that wants to learn is frustrated but the other part that knows your telling me this route for a good reason is going "if i must"...  I take it the process to get what i am looking for is rather complex? i think i have to apply filters at some point also. part of my confusion also is all these demos seem to require more than one point in time to make this work... so when im trying to look at just timestamp 13:25:41.160 and everything else wants multiple.. its only adding to my confusion... 
  • wjcroftwjcroft Mount Shasta, CA
    Did you try the mouse / menu actions suggested in these similar links to drag the off screen window back onto your main screen?


    I have no idea why BrainBay is doing this, I've only ever used it on single monitor machines. Obviously a bug of some sort, but you may be able to work around. For example some of these pages from the search suggest right clicking on the bottom of screen icon, then select Move and use the arrow keys to move the window back to visible. Or Tile, etc.

    re: tutorial, you actually do not need to run the demos to understand how it explains the signal processing that happens in the element blocks. You can look online for various signal processing tutorials. The raw EEG file or data stream is essentially a time series of microvolt values at the 250 hz sample rate. The first filter (say a .5 hz highpass), would remove the DC offset. You can then apply various bandpass filters, for example 8-12 hz for alpha, and then turn that (absolute values) into a band amplitude, averaged over a period of time, say a half second. Other element blocks can take band amplitudes and generate feedback based on that.

    The advantage of VPL programs such as BrainBay, BioEra, Bioexplorer, OpenViBE, etc. -- is that you can do all this signal processing without any programming languages. If you ditch the VPL, then you have to call functions to do each of those steps. And since the functions need to operate on continuous data streams, the programming of DSP can be complicated. VPLs make it easy, since the function blocks connect directly their appropriate input stream, to output stream, with a DSP operation between.

  • MitchMMitchM Appleton, WI, USA
    hmm, yes I can see why it would be better to use these. still id hate to have to lug my laptop around when you have a phone or tablet that could assist with recording the data on the go (driving walking) it create a bit of artifact noise as well, but it was interesting to do the SRB2 SRB1 & Cz configuration.. also the way i am envisioning what i am developing, I'm trying to stay away from the live visual charting/plot graph stuff and use the mobile to configure the parameters and basically display a numeric value that keeps getting updated for each channel, like what the "Band Power" widget does in the GUI, then let the lights to the work and sound as well... with session performance data at the end. I don't see many demos with iOS devices and mostly android... i wonder if there is any hardware limitation factors there (besides provisioning profiles, etc)
  • wjcroftwjcroft Mount Shasta, CA
    BioEra supports Android,


    You'd need an OTG compatible device to support the Cyton dongle. The Cyton dongle speaks non-standard Bluetooth, so is not compatible with Bluetooth 2.x, 3.x, or 4.x (BLE). Since BioEra supports FTDI serial port devices (which Cyton dongle is), it should extend to Cyton (although not currently listed on above link).

    Wifi connection to the mobile might work, but not currently with BioEra.

    Neurofeedback generally requires a seated position to avoid artifacts from EEG cable motion / different pressures on the electrodes. 
  • MitchMMitchM Appleton, WI, USA
    that is what I saw when i was doing my initial research before buying... currently im having issues with the wifi shield which i am told AJ will connect with me at some point on..

    With my laptop connected to the wifi shield and nothing else the following things happen

    1. Search fails to find it. (blue shield blinked 10 times and my pc connected to it)
    2. “Static IP” finds it but only using UDP and only using the default settings
    3. Or if it connects and OPEN BCI launches, I get not data from it… with the Cyton in “pc” mode or in “off” mode. trying different speeds and sample rates does not return anything either. I have the latest firmware and have not clicked the "update firmware button"

    No idea what’s going on as I tried it in different locations thinking some of the wifi signals around me were causing interference and I don’t think it was designed to be this difficult.

  • wjcroftwjcroft Mount Shasta, CA
    AJ is your guy. You can try PMing him.
  • MitchMMitchM Appleton, WI, USA
    Just wanted to leave a message on here that after I reformatted my computer, I was able to get BrainBay to open properly. have not yet tested it with my Cyton yet, but I figured id pass the message along as today I will be attempting the tutorials you mentioned above. I am going to see if for some reason the format fixed the wifi shield issues also.. will post here with an update...


  • Let me ask the other questions about neurofeedback using OpenBCI.

    I'm going to create a neurofeedback software using OpeBCI.
    I had already developed a woking protocol using Emotiv, so I want to port it to OpenBCI.
    I know that I should make a completely different software suit for OpenBCI, and it is not a problem because I'm a software developer.

    My questions are:
    - Can I use OpenBCI from my Mac?(It it better if I can use Wifi to connect the OpenBCI board from my Mac)
    - Are there software SDK or library to capture the realtime EEG raw data?(I prefer to use python)
    - I have to calculate ERP, so it is better to synchronize the timing of stimuli and EEG, and I want to send a trigger to OpenBCI. Is it possible? and how?

    I tried to find the reference on this site that describes the functions to get EEG raw data, and the format of the data, but I couldn't.
    I found these documents, but I cannot have a confidence that I can create a my neurofeedback software...
  • wjcroftwjcroft Mount Shasta, CA
    Hiroki, hello.

    The Cyton is easy to interface with Python, using the OpenBCI_Python repository as you have found, which works on all platforms (Mac, Linux, Windows).


    You DO NOT need to use the SDK commands yourself (in most cases), the Python interface does most of the work for you.

    There is a search box and button on the right column, and a number of threads on external triggers, ERP, etc.

    Regards,

    William

  • Hi William,

    Thank you for the explanation.
    I'm going to purchase the Cyton Board, pro-assembled Ultracortex headset, and dry EEG electrodes.
    I'll buy two lithium ion batteries and a charger from adafruit as you recommend.

    Are there any other items I should purchase from you in order to do a real-time neurofeedback experiment?
  • Hi,

    The OpenBCI EEG Headband Kit is relatively inexpensive.
    I'm happy if I can use it instead of the Ultracortex headset.
    I want to measure EEG at Fz, but the product description doesn't include Fz.
    I also want to use dry electrodes.
    Is it usable for the measurement at Fz using dry electrodes?

  • wjcroftwjcroft Mount Shasta, CA
    Hiroki, hi.

    The kit you mention,


    Does contain the spiky electrodes used at scalp locations with hair. However the velcro band included is used in the headband position. You need a "cross strap" that bridges across the left right sides to cover the top of the head. This cross strap will hold your Fz or F3 or F4 electrodes.

    The velcro band material is simply called Velcro One-Wrap in 3/4 inch width,


    You will need to punch a hole at locations you want for the electrodes. Can be done with a hole punch or drill bit.

    Regards,

    William

    PS more of these dry / velcro supplies at FRI,

  • Hi William,

    I purchased the Cyton Board and pro-assembled Ultracortex headset.
    I confirmed they work as expected using OpenBCI_GUI.
    Thank you for your help.

    Now I'm going to create my own program using python.
    But I have problem with them.

    First, the python code written here works:
    Executing board.start_streaming(handle_sample) prints tuples of 8 samples.
    But these values are different as expected, like -910.69947 and -16807.148373.

    I want to know the micro voltage of the measured EEG.
    How to convert these variables to uV?

    Next, I tried to find the way to send an external trigger from python as you suggested, but couldn't.
    Can you indicate the reference?

    Finally, the Cython seems to have a hardware notch filter.
    Using OpenBCI_GUI, I can apply 50Hz notch filter.
    How can I apply this 50Hz notch filter from python code?
    Sending 'f' to serial port seems to activate the filter, but I want to apply 50Hz, not 60Hz.

    with best regards,
Sign In or Register to comment.