BioSemi.nl

Hello Biosemi team,
I wish to know if there are detailed reviews that I can read to properly understand the concept of offset as it applies to EEG acquisition.

The questions that I have are as follows:
1. What is the definition of offset?
ActiveTwo operation manual says it is the "running average of the voltage measured between CMS and each active electrode" and that it "reflects the half-cell potential of the electrode/gel/skin interface." I wish to understand what these statements mean.
2. What is the relationship between offset and data quality/electrode noise?
Forum readings suggest that while low offset generally indicates low electrode noise due to good contact between the skin/ge/electrode, it is not a 1:1 relationship. I am wondering if there are any previous research articles or reviews that investigated this relationship.
3. What is the significance of +/- 40 mV offset criteria?
4. I understand that offsets need to be within +/- 40 mV for optimal electrode/gel/skin contact interface but what does low offset actually represent in terms of this contact interface? Does it mean charge can flow better?
5. Why is it that if you are using an active electrode system one does not require impedance to be measured?
6. Are there any other EEG acquisition devices that also use offsets instead of impedance?

Sorry for the bombardment of questions.
Many thanks in advance for your time.

Following the above post, I found an article that would be interesting to the audience in this field:
https://www.neuroregulation.org/article/view/14604
The article compares DC offset and impedance readings in electrode connection quality and concludes that DC offset is not a reliable measure of electrode connection quality and that there is no linear relationship between offset and impedance.

1. What is the definition of offset?

The electrode-gel-skin interface forms a galvanic half-cell. The half-cell potential is determined by the electrode metal. The offset voltage is the potential difference between two of these half-cells. When the electrodes are identical (same metal and perfectly pure), the half-cell potentials are the same, and the offset voltage between the electrodes is zero. The offset voltage between a silver and a gold electrode is not zero because the different metals generate different half-cell potentials. Pollution of an electrode affects the half-cell potential. So, an Ag-AgCl electrode that is polluted with traces of non-silver metal will have an offset with respect to a pure Ag-ACl electrode. In addition, pollution of the electrode will generate an unstable half-cell potential, which is seen as (pink) electrode noise

2. What is the relationship between offset and data quality/electrode noise?

High offsets indicate electrode pollution which generally generates electrode noise

5. Why is it that if you are using an active electrode system one does not require impedance to be measured?

A resistor generates thermal noise, see https://en.wikipedia.org/wiki/Johnson%E ... uist_noise . However, with typical EEG electrodes, the instability of the electrode half-cell potential is usually the dominant noise source (thermal noise begins to play a role only for impedances of several hundreds of kOhm). The 5k limit used with passive electrodes originates from the problem that high differences in electrode impedance generates (50/60 Hz) interference, see https://www.biosemi.com/publications/pd ... uction.pdf . With active electrodes, the source impedance is transformed to less than 1 Ohm on the electrode, and no interference currents flow through the electrode-gel-skin interface. Note that with some skin scrubbing, it is no problem to achieve a low impedance for a polluted electrode with high offset and noise.

7. The article compares DC offset and impedance readings in electrode connection quality and concludes that DC offset is not a reliable measure of electrode connection quality and that there is no linear relationship between offset and impedance.

The paper argues that high impedances cause interference problems with passive electrodes, and that these high impedances can not be identified by looking at the offsets. However, the point is that high electrode impedances do not cause interference problems with active electrodes, and that excessive electrode noise can not be identified by looking at the electrode impedance.

Best regards, Coen (BioSemi)

Thank you for the reply Coen,
"With active electrodes, the source impedance is transformed to less than 1 Ohm on the electrode, and no interference currents flow through the electrode-gel-skin interface."
Can you recommend any journal articles that I can cite for this sentence?
I assumed the article you referenced contained information regarding this, but I had trouble finding it. (https://www.biosemi.com/publications/pd ... uction.pdf)

Google on "opamp buffer output impedance". With active electrodes, the electrode impedances in the interference paper (Ze in Fig. 1) are reduced from tens of kOhm to less than an Ohm, and the magnitude of interference is reduced with the same ratio.

Best regards, Coen (BioSemi)