Scaling and Labelling issues

[xaaronx]

Hi,

I have encountered a couple issues with recording whilst using the LSL and an ANT eego device.

The data records as it should but the labelling of the channels is rather peculiar. They are labelled "microvolts" or "acquisition" or " serial number" etc.. Have you any idea why that might be?

Also the data that we have recorded seem to be scaled incorrectly (the voltages are far lower than expected). They are too low to be realistic and so we assume that they have simply been scaled incorrectly. Is this something that has been reported ever?

Please let me know,

Aaron

[dmedine]

Is there any way you can send me the data you recorded so I can check it out (I'm at dmedine'at'ucsd'dot'edu)? I don't have one of these devices to test with, but a quick look at the meta-data generation code tells me it is correct (which by no means indicates that it actually is correct...).

It should be that each channel node has three children in the meta-data, one for "label" (which should be "Ch1", "Ch2", etc.) one for "label" which should be "EEG" for every channel, and one for "unit" which should be "microvolts" for each channel.

In addition, there is node in the meta-data called "acquisition" which has child values for "manufacturer" which should be "antneuro" and "serial_number" which should be the device serial number.

I don't see any manipulation of data in the code itself, so I haven't any guesses as to where your scaling issue is arising. Perhaps looking at the data will help.

On 6/22/2016 10:39 AM, xaaronx wrote:

Hi,

I have encountered a couple issues with recording whilst using the LSL and an ANT eego device.

The data records as it should but the labelling of the channels is rather peculiar. They are labelled "microvolts" or "acquisition" or " serial number" etc.. Have you any idea why that might be?

Also the data that we have recorded seem to be scaled incorrectly (the voltages are far lower than expected). They are too low to be realistic and so we assume that they have simply been scaled incorrectly. Is this something that has been reported ever?

Please let me know,

Aaron

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/sccn/labstreaminglayer/issues/123, or mute the thread https://github.com/notifications/unsubscribe/ADch7pKOT-bs8xHcyM-aV7kh7GJ6t1_2ks5qOXNkgaJpZM4I8CSp.

[xaaronx]

Hi there,

I have written to you personally because I would prefer not to share the data publicly. Did you receive the email?

Thanks!

[atom2626]

Hi there,

I did get the email, but I doubt your wish was to write to me.

Fred

---

From: xaaronx [email protected] Sent: Monday, June 27, 2016 08:24 To: sccn/labstreaminglayer Subject: Re: [sccn/labstreaminglayer] Scaling and Labelling issues (#123)

Hi there,

I have written to you personally because I would prefer not to share the data publicly. Did you receive the email?

Thanks!

You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/sccn/labstreaminglayer/issues/123#issuecomment-228730720, or mute the threadhttps://github.com/notifications/unsubscribe/AKzynDpzMeTnFIACeXJbryJ-JWTPqCsqks5qP8DzgaJpZM4I8CSp.

[dmedine]

I have been doing a little checking into the eego amp code and I have found that there is indeed a bug with the stream meta-data code. I can test (and thus fix) this, but since I don't have an eego device or the premium sdk right now, I'm not in a position to address the other issue (scaling of eeg data).

For now, I can fix the labeling issue and I suppose we will go from there. I will ask around to see if I can get my hands on a device. In the meantime, are you in a position to verify the data with a test signal---i.e. send one channel a 500 microvolt, 100 Hz sinusoid and see what it looks like in the recorded data?