Bad connection

Se7en
Posts: 33
Joined: Wed May 08, 2019 10:26 pm

Bad connection

Post by Se7en »

Hello James,

I usually check my recoding with neutovisual as.it.shows bad connection for each sensor in %
Is there an easy way to include that data into your website?
I have perfectly fine graphs that show no BF/GF and still have up to 5% connection errors.

Obviously that could influence the graph significantly.

Thanks for considering
User avatar
James
Site Admin
Posts: 1110
Joined: Wed Jan 02, 2013 9:06 pm

Re: Bad connection

Post by James »

Switch to the raw view to check signal quality.

The smaller the deviation from the mid point the better the signal.
Se7en
Posts: 33
Joined: Wed May 08, 2019 10:26 pm

Re: Bad connection

Post by Se7en »

Amazing information.
When is the trigger point for a BF?

Sven
User avatar
James
Site Admin
Posts: 1110
Joined: Wed Jan 02, 2013 9:06 pm

Re: Bad connection

Post by James »

I'm not sure sorry. This is generated by Interaxon's code which they do not share.
Se7en
Posts: 33
Joined: Wed May 08, 2019 10:26 pm

Re: Bad connection

Post by Se7en »

Kind of sad that they don't open up more to let other developer develop interesting application for a truly amazing device like.muse.

Thanks for the help and thanks for the app
Se7en
Posts: 33
Joined: Wed May 08, 2019 10:26 pm

Re: Bad connection

Post by Se7en »

James I did some "tests". I have a graph with some BF/GF
About 6 BF over a 40 min session.
I was expecting the graph set on all data points and raw to have spikes at the BF points but the graph showed an overall thicker line and no spikes.
Neurovisual shows a 2% connection loss on all 4 sensor.

Therefore I guess a BF is a loss of contact to all 4 sensor at the same time.

In my opinion that would mean the entire recording is compromised and not accurate.

Is that correct?
User avatar
James
Site Admin
Posts: 1110
Joined: Wed Jan 02, 2013 9:06 pm

Re: Bad connection

Post by James »

If you want to check signal quality in this way you will need to record all the raw data by setting the recording interval to Constant.

You will only see spikes when you blink. Signal quality will be the difference from the normal.
Se7en
Posts: 33
Joined: Wed May 08, 2019 10:26 pm

Re: Bad connection

Post by Se7en »

That's going to be a huge file 🙏🙏🙏
User avatar
James
Site Admin
Posts: 1110
Joined: Wed Jan 02, 2013 9:06 pm

Re: Bad connection

Post by James »

Precisely why the default is 1hz ;-)
Se7en
Posts: 33
Joined: Wed May 08, 2019 10:26 pm

Re: Bad connection

Post by Se7en »

James
do I understand right that the scale on raw+all data points is in microVolts?

What would be in your opinion an acceptable diviation from the midpoint?
My midline is usually between 820+880 so an up and down of 60 microVolts.
Post Reply