EDA: Signal in the data sheet and positioning of the electrodes

Connection, ECG, EMG, EDA, etc.
TEL
Posts: 8
Joined: Thu Apr 13, 2017 8:04 am

EDA: Signal in the data sheet and positioning of the electrodes

Postby TEL » Tue Apr 18, 2017 11:48 am

Dear Bitalino community,

I am a computer science student who is working on a sensor device using the Bitalino EDA 151015.

I: Signal in the data sheet
For the first step I would like to reproduce the signal given in the data sheet. On the Bitalino Admins already gave me the information that the vasalva maneuver was used for this. As far as I understood the maneuver on closes his/her own nose and mouth and then exhales intensively for a period of 10 seconds. While doing this a doctor would check your ears.
As the signal is shown over a period of 100 seconds I would like to know if there was a repetition of the maneuver, if there were some rest periods and (if possible) the timestamps for starting and/or beginning of each maneuver.

II: Positioning of the electrodes
The data sheet mentions the thenar eminence as electrode positions. Also the index finger and middle or third finger are mentioned here: viewtopic.php?f=3&t=135

Poh et al. (http://web.media.mit.edu/~zher/papers/Poh-etal-TBME.pdf) mentioned the wrist as a valid position for recording eda. It is also mentioned that to detect EDA a current is applied to the skin beneath the two electrodes. Now I have two questions:
1. Is it possible to detect EDA using the Bitalin EDA 151015 with electrodes placed at the wrist as shown in the paper by Poh et al.?
2. How does the Bitalino EDA 151015 apply current to the skin to detect EDA?

Thank you very much, I am curious to read some answers.

BITalino
Site Admin
Posts: 567
Joined: Tue Aug 27, 2013 3:47 pm

Re: EDA: Signal in the data sheet and positioning of the electrodes

Postby BITalino » Tue Apr 18, 2017 7:23 pm

Hi there,

First of all we'd like to thank you for your contact and preference for BITalino. As for your questions:

1) The manoeuvre was indeed repeated (hence the multiple bumps you see), but it was done in a highly informal way just to provoke a reaction and without having a doctor doing any procedure. The best to have accurate information regarding the stimuli / reaction timings is to replicate an experimental setup of your own. Just a small detail... the time scale on the graphic is in milliseconds, and as such the duration of the signal shown there is around 20 seconds.

2) With the standard BITalino sensor in principle from the wrist placement it may not be easy to extract meaningful information. We have a modified version that has better performance in this location using dry electrodes for example. As for the measurement principle, the sensor applies a small (known) voltage on the skin and has a set of feedback resistors of known values... the measurements between the sensing leads will vary with the changes in the skin moisture (or resistance), which are amplified to provide higher sensitivity resulting in the output of the sensor. If you want to perform a deeper analysis send us an e-mail and we'll share the schematics.

For more information regarding the EDA, measurement principles and other useful details, please refer to:
http://www.springer.com/us/book/9781461411253
http://www.bem.fi/book/27/27.htm

Best regards,
The BITalino Team

TEL
Posts: 8
Joined: Thu Apr 13, 2017 8:04 am

Re: EDA: Signal in the data sheet and positioning of the electrodes

Postby TEL » Thu Apr 20, 2017 2:30 pm

Thank you very much for your reply so far.

In 1) I was referring to this data sheet: http://bitalino.com/datasheets/EDA_Sensor_Datasheet.pdf
The whole duration of it is 100 seconds right? Which data sheet did you mean with the 20 seconds signal? I am interested in getting as much information (example graphs and/or data) as possible for comparing the output of my sensor device with it.

2) Your modified version sounds very interesting as I plan to use dry electrodes. Where is the difference to the usual one? How can I get the modified version?

Another question came up when I did some further reading. Dr Braithwaite et al. (http://www.birmingham.ac.uk/Documents/c ... tivity.pdf) name 35 Hz as the upper border of the occurring frequencies. I am pretty sure you have a very good reason for choosing 3 Hz. My professors will most likely ask why I decided to use a sensor with a filter at 3 Hz so could you share this reason with me?

Thank you very much, I am really looking forward to seeing the answers.
Tom

BITalino
Site Admin
Posts: 567
Joined: Tue Aug 27, 2013 3:47 pm

Re: EDA: Signal in the data sheet and positioning of the electrodes

Postby BITalino » Fri Apr 21, 2017 12:02 am

Hi Tom,

Regarding 1) the data sheet you're looking at may be a legacy version that has been discontinued in the meanwhile... please check the physical appearance of your sensor and see if the following data sheet is the correct one (and yes, in that version of the data sheet its showing 100 seconds):
http://bitalino.com/datasheets/REVOLUTION_EDA_Sensor_Datasheet.pdf

As for 2) the difference is a change in the gain that makes the sensor more sensitive; we can tell you which components to replace (although those will be SMD), but its always easier and more convenient to simply order the modified sensor from us... should you choose to purchase, these are the items:
https://store.plux.info/bitalino-sensors/11-electrodermal-activity-eda-sensor.html
https://store.plux.info/support-packs/67-advanced-support-1-h-820202305.html

In what concerns the passing band on the filter it can also be changed, 35 Hz is the theoretical upper frequency (e.g. used in BioPac), however there is a heterogeneous ecosystem of options (e.g. Shimmer uses 15.9 Hz).

Despite this diversity, several references seemed to point towards an empirical limit between 2-5 Hz (e.g. the Bioelectromagnetism book by Malmivuo & Plonsey - linked bellow - Chapter 27), and as such we designed the low pass filter to have a cut-off frequency that sits within said interval:
http://www.bem.fi/book/

Best regards,
The BITalino Team

TEL
Posts: 8
Joined: Thu Apr 13, 2017 8:04 am

Re: EDA: Signal in the data sheet and positioning of the electrodes

Postby TEL » Fri Apr 21, 2017 12:27 pm

Thank you very much, this large amount of information helps me and will keep me busy for the next weeks.

One more question that is not related to my student's but came up by my professor. He wanted to know if you have any partners or company-recommendations for assembling devices using Bitalino Sensors (for example using a microprocessor (eg. STM Nucleo), bluetooth module, Bitalino sensors to create a circuit, assemble the device and put a case around it). In an optimal case this parter is based in Berlin, Germany.

Anyhow,
thank you very much for your help!

BITalino
Site Admin
Posts: 567
Joined: Tue Aug 27, 2013 3:47 pm

Re: EDA: Signal in the data sheet and positioning of the electrodes

Postby BITalino » Fri Apr 21, 2017 12:43 pm

Hi there,

We're glad to have been of assistance. One possible contact point in Germany is the team at the Makery:
http://makery.de

In particular we've met before with Martin Poggenclaas, but they're based out of Hamburg.

If you have an assembly in mind, and if you're interested, we may be able to offer that service also.

Best regards,
The BITalino Team

TEL
Posts: 8
Joined: Thu Apr 13, 2017 8:04 am

Re: EDA: Signal in the data sheet and positioning of the electrodes

Postby TEL » Tue Apr 25, 2017 11:58 am

Thank you very much, we are about to contact the Makery.

We are thinking about ordering two full kits to have devices to start with. Therefore we need to know the following:
Is it possible to detect EDA and ECG together?

BITalino
Site Admin
Posts: 567
Joined: Tue Aug 27, 2013 3:47 pm

Re: EDA: Signal in the data sheet and positioning of the electrodes

Postby BITalino » Tue Apr 25, 2017 2:33 pm

We're always glad to be of assistance.

For sure you can acquire ECG and EDA together... in fact you can record up to 6 analog inputs simultaneously.

If you have further questions let us know.

Best regards,
The BITalino Team


Return to “Sensors”




cron