Integrating the 5 channels to get just one brain wave

begozmen
Posts: 4
Joined: Wed Jul 05, 2017 5:15 pm

Integrating the 5 channels to get just one brain wave

Post by begozmen »

Hi,
I am wondering how can I get the single wave that I see in the muse monitor app but in my computer. I have read in the technical manual that: 'Brain wave values are absolute band powers, based on the logarithm of the Power Spectral Density (PSD) of the EEG data for each channel'.
So if I use the pwelch function in Matalb it gives me the power density function of each channel but i don't know how to integrate them.
Thank you
Begonya
User avatar
James
Site Admin
Posts: 1111
Joined: Wed Jan 02, 2013 9:06 pm

Re: Integrating the 5 channels to get just one brain wave

Post by James »

Sorry, I've not used MatLab, so I can't give you any specifics.

The single absolute values are just averages of the absolute values for each of the four channels: (TP9+AF7+AF8+TP10)/4.
begozmen
Posts: 4
Joined: Wed Jul 05, 2017 5:15 pm

Re: Integrating the 5 channels to get just one brain wave

Post by begozmen »

Thank you for your response.
But, they are averages for the absolute values that come directly in the file of muse monitor or the averages of the logarithm of the PSD (I got this information from the technical manual of musemonitor (see attachment)
Attachments
absolute.PNG
absolute.PNG (97.34 KiB) Viewed 6942 times
User avatar
James
Site Admin
Posts: 1111
Joined: Wed Jan 02, 2013 9:06 pm

Re: Integrating the 5 channels to get just one brain wave

Post by James »

In the screenshot above, what you are seeing is the average of the four sensor values, which are output in the recording file.

You can also see the individual values on screen by going in to Settings and selecting what you want under "Wave Types (Absolute)" where you can pick from "Average Values", "Left/Right", "Front/Back" and "All Sensors".
begozmen
Posts: 4
Joined: Wed Jul 05, 2017 5:15 pm

Re: Integrating the 5 channels to get just one brain wave

Post by begozmen »

Bu if i consider the alpha channel in the app I see a range of 60-90 dB but if i plot the same values in matlab as you said(taking the alpha absolute values), the range is from 0 to 1. Even if i convert the absolute values to PSD and take the log, it is still in 0-1 range.
How did you get a 60-90 range?
Thanks
User avatar
James
Site Admin
Posts: 1111
Joined: Wed Jan 02, 2013 9:06 pm

Re: Integrating the 5 channels to get just one brain wave

Post by James »

As per the online documentation:
The EEG PSD values as read from the sensors are commonly in the {-1:+1} range. For the text value display these are shown in a more legible {0:100} range.
display_value = (value+1)/0.02
Henke
Posts: 8
Joined: Sun Oct 29, 2017 3:54 am

Re: Integrating the 5 channels to get just one brain wave

Post by Henke »

Hi,

A follow-up of this topic.
I can see from this post that average values are calculated as logaritms (not in linear units). I can verify that once I have valid values in all four channels. But very often there is "noise". What does channel values of 0 mean and what does a large (8 digit) number means and how are they handled in the average?

Thanks :-)
User avatar
James
Site Admin
Posts: 1111
Joined: Wed Jan 02, 2013 9:06 pm

Re: Integrating the 5 channels to get just one brain wave

Post by James »

What 8 digit numbers are you talking about? Band powers are usually {-1:+1}. The highest I've seen is only around +5.
Henke
Posts: 8
Joined: Sun Oct 29, 2017 3:54 am

Re: Integrating the 5 channels to get just one brain wave

Post by Henke »

Sorry my mistake. Excel import delimiter problem solved.
Thanks :-)
LinDN
Posts: 13
Joined: Tue Jul 25, 2023 4:33 am

Re: Integrating the 5 channels to get just one brain wave

Post by LinDN »

James wrote: Thu Jul 06, 2017 3:00 pm As per the online documentation:
The EEG PSD values as read from the sensors are commonly in the {-1:+1} range. For the text value display these are shown in a more legible {0:100} range.
display_value = (value+1)/0.02
The value of Theta_TP9 shown in my csv record file is 1.462454573. I would like to ask you a few questions about this. 1. Why is it higher than 1? 2. According to the instructions, its unit should be bels. But 1 bels = 10dB (I'm not sure if my knowledge is correct), 14.6dB obviously should not be the normal value for alpha waves. If we deduce the reverse according to your calculation formula, the corresponding display value of 1.462454573 is 123, it is more likely to be the dB value of alpha wave, but I do not know why I want to calculate this way, I hope you can answer it for me, thank you.
Post Reply