Page 1 of 2
Preprocessing raw data
Posted: Fri May 03, 2019 3:47 am
by ty22
Hi,
I really like your app thanks!
I'm some questions:
1. how do you preprocessing the data to remove the artefacts?
2. if I have a piece of recorded data and I want to know if the person is in a highly nervous/frightened state, should I compute the average value of all 4 electrodes for Beta band or is there any other ways?
Thanks!
Jeanne
Re: Preprocessing raw data
Posted: Fri May 03, 2019 9:03 am
by James
1 - I don't. It's done automatically by the Interaxon SDK before my code gets the values. I believe they are removing blinks and power line interference.
2 - I'm not sure sorry. My personal research has been focused on relaxation.
Re: Preprocessing raw data
Posted: Sat May 04, 2019 11:01 pm
by ty22
James wrote: ↑Fri May 03, 2019 9:03 am
1 -
I don't. It's done automatically by the Interaxon SDK before my code gets the values. I believe they are removing blinks and power line interference.
2 - I'm not sure sorry. My personal research has been focused on relaxation.
Thanks for your reply, I've read many articles on how brain works, about frequency, DSP and so on.
But it is still difficult for me to understand the data from Muse monitor. So could you tell me about what kind of brain activity can be seen on this picture? Does the Session Average Relative tell anything about asymmetry?
link:
https://imgur.com/a/93JWB0D
Thank you so much!
Re: Preprocessing raw data
Posted: Sun May 05, 2019 10:43 am
by James
All I can tell from this is that the subject wasn't relaxed. Sorry, like I said, my personal research is on relaxation!
Regarding symmetry, the little graph in the bottom left, shows the left vs right power for each wave.
In this example you can see it was more left than right. If you hover your mouse over the individual bars it'll show you the actual numbers.
Re: Preprocessing raw data
Posted: Sun May 05, 2019 2:11 pm
by ty22
James wrote: ↑Sun May 05, 2019 10:43 am
All I can tell from this is that the subject wasn't relaxed. Sorry, like I said, my personal research is on relaxation!
Regarding symmetry, the little graph in the bottom left, shows the left vs right power for each wave.
In this example you can see it was more left than right. If you hover your mouse over the individual bars it'll show you the actual numbers.
Thanks James! Could you also tell me how this bottom left graph is computed? Is it averaging each frequency band from all four electrodes? Why is the left one always negative? Thanks!
Re: Preprocessing raw data
Posted: Sun May 05, 2019 4:33 pm
by James
The bottom left graph is the average of the left and the average of the right for each band over the entire session.
If you click on a specific point on the graph, you will get another graph shown under it showing just that point in time.
The left side is always negative so that it renders correctly with the blue bars on the left and the red bars on the right.
Re: Preprocessing raw data
Posted: Mon May 06, 2019 8:55 pm
by ty22
James wrote: ↑Sun May 05, 2019 4:33 pm
The bottom left graph is the average of the left and the average of the right for each band over the entire session.
If you click on a specific point on the graph, you will get another graph shown under it showing just that point in time.
The left side is always negative so that it renders correctly with the blue bars on the left and the red bars on the right.
Thank you for your reply! Do you know if subject's emotional reaction to sudden visual stimuli (for example, a terrifying face) could be reflected on the Alpha (or Beta) change, or frontal asymmetry values, right after the onset of stimuli?
Thank you very much!
Re: Preprocessing raw data
Posted: Tue May 07, 2019 8:14 am
by James
Probably, but I'm not sure what it would look like.
Re: Preprocessing raw data
Posted: Fri May 10, 2019 11:33 am
by ty22
James wrote: ↑Tue May 07, 2019 8:14 am
Probably, but I'm not sure what it would look like.
Hi James, since the voltage of raw EEG signal usually lies between 850+-100uV, I see a lot of outliers due to lack of contact of sensors during the tests. Do you think I can remove some noise by deleting everything out of this range?
Re: Preprocessing raw data
Posted: Fri May 10, 2019 11:52 am
by James
Sure.