Our school is considering a bid for an EEG rig, and I like the look of Biosemi. I like the idea of locally amplified electrodes and am a bit surprised that this hasn't been a standard feature of all EEG setups.
Shielding is something of a black art, but I'm wondering what your experiences are, specifically, to what degree do you think it's necessary to build shielding into a room, particularly using your active electrodes?
-Brad Wyble
University of Kent, Canterbury
Shielding in EEG
-
Coen
- Site Admin
- Posts: 1152
- Joined: Fri Mar 26, 2004 7:00 pm
- Location: Amsterdam, Netherlands
- Contact:
Shielding in EEG
A Faraday cage is used to achieve two things: elimination of interference currents in the electrode wires, and elimination of interference current in the ground electrode (DRL electrode in case of the ActiveTwo). Both mechanisms cause 50 Hz noise in the data, it is anything but a black art, as you can read in http://www.biosemi.com/publications/pdf ... uction.pdf.
In the ActiveTwo system, the effect of interference currents in the electrode wires is practically eliminated by using active electrodes. The active electrode has an output impedance of less than 1 Ohm. So, any interference current in the electrode wire, flows through an impedance of a factor 10,000 to 100,000 smaller than in the case of a passive electrode. Consequently, the resultant interference voltages are decreased by this factor.
In the ActiveTwo system, the interference currents through the ground electrode are minimized by using battery power supply for the front-end, and by using fiber optic data transfer from the front-end to the PC. This setup offers a near perfect galvanic isolation between front-end and environment. This results in minimal stray capacitances, and minimal current between front-end and subject. The resultant stray capacitances are a factor 10-100 smaller than can be achieved with the best insolation transformers (power transfer) and optocouplers (data transfer). Minimal current via the Ground (DRL) electrode means minimal Common Mode voltage (CM is the product of Ground electrode impedance and current flowing via this electrode). In addition, minimal current via the Ground electrode result in minimal interference currents trough the subject body. This means a decrease of differential inference voltage across the subject surface, see http://www.biosemi.com/faq/battery_and_fiber.htm
The unique setup of the ActiveTwo already handles both functions of a cage of Faraday. A cage of Faraday will therefore not further improve the quality of the data. In fact, the majority of ActiveTwo users apply the system successfully outside a Faraday cage.
In the ActiveTwo system, the effect of interference currents in the electrode wires is practically eliminated by using active electrodes. The active electrode has an output impedance of less than 1 Ohm. So, any interference current in the electrode wire, flows through an impedance of a factor 10,000 to 100,000 smaller than in the case of a passive electrode. Consequently, the resultant interference voltages are decreased by this factor.
In the ActiveTwo system, the interference currents through the ground electrode are minimized by using battery power supply for the front-end, and by using fiber optic data transfer from the front-end to the PC. This setup offers a near perfect galvanic isolation between front-end and environment. This results in minimal stray capacitances, and minimal current between front-end and subject. The resultant stray capacitances are a factor 10-100 smaller than can be achieved with the best insolation transformers (power transfer) and optocouplers (data transfer). Minimal current via the Ground (DRL) electrode means minimal Common Mode voltage (CM is the product of Ground electrode impedance and current flowing via this electrode). In addition, minimal current via the Ground electrode result in minimal interference currents trough the subject body. This means a decrease of differential inference voltage across the subject surface, see http://www.biosemi.com/faq/battery_and_fiber.htm
The unique setup of the ActiveTwo already handles both functions of a cage of Faraday. A cage of Faraday will therefore not further improve the quality of the data. In fact, the majority of ActiveTwo users apply the system successfully outside a Faraday cage.
-
Coen
- Site Admin
- Posts: 1152
- Joined: Fri Mar 26, 2004 7:00 pm
- Location: Amsterdam, Netherlands
- Contact:
What exactly do you want to measure: the Isolation Mode Rejection Ratio (IMRR), the isolation capacitance, the breakdown voltage of the isolation, or some other parameter ?
If you want to contribute to this forum, I would appreciate it if you would at least put some effort into writing a clear question.
Coen (BioSemi)
If you want to contribute to this forum, I would appreciate it if you would at least put some effort into writing a clear question.
Coen (BioSemi)
[quote="Coen"]What exactly do you want to measure: the Isolation Mode Rejection Ratio (IMRR), the isolation capacitance, the breakdown voltage of the isolation, or some other parameter ?
If you want to contribute to this forum, I would appreciate it if you would at least put some effort into writing a clear question.
Coen (BioSemi)[/quote]
Sorry, I mean Isolation Mode Rejection Ratio (IMRR), the isolation capacitance and input impedance. I want to find these figures for my EEG device, exactly. I think, chip impedance ressitance maybe one argument, but the input impedance must be changed with frequency (0-70 Hz). Could you please give me information about these measurement and calculations, regards.
If you want to contribute to this forum, I would appreciate it if you would at least put some effort into writing a clear question.
Coen (BioSemi)[/quote]
Sorry, I mean Isolation Mode Rejection Ratio (IMRR), the isolation capacitance and input impedance. I want to find these figures for my EEG device, exactly. I think, chip impedance ressitance maybe one argument, but the input impedance must be changed with frequency (0-70 Hz). Could you please give me information about these measurement and calculations, regards.
-
Coen
- Site Admin
- Posts: 1152
- Joined: Fri Mar 26, 2004 7:00 pm
- Location: Amsterdam, Netherlands
- Contact:
IMRR: apply test signal with known voltage across the isolation barrier, and measure the effect on the system's output signal (see http://www.biosemi.com/publications/pdf ... ection.pdf)
Isolation capacitance: apply (AC) test signal with known voltage across the isolation barrier, and determine the current
Input impedance: determine the decrease in CMRR when imbalanced source impedances are applied (see http://www.biosemi.com/publications/pdf ... uction.pdf)
Best regards, Coen (BioSemi)
Isolation capacitance: apply (AC) test signal with known voltage across the isolation barrier, and determine the current
Input impedance: determine the decrease in CMRR when imbalanced source impedances are applied (see http://www.biosemi.com/publications/pdf ... uction.pdf)
Best regards, Coen (BioSemi)
EEG in close proximity to an MR scanner
Does anyone have experience setting up an EEG lab nearby to an MRI scanner. I am wondering what the possible issues are (both for Biosemi's active electrodes and other EEG systems). Without the EEG rig set up, how would one go about testing for whether there will be electromagnetic noise from the magnet? Of course the magnet is in a shielded room, but the proposed location for an EEG rig is less than ~100 feet away. Should I worry?
-
Coen
- Site Admin
- Posts: 1152
- Joined: Fri Mar 26, 2004 7:00 pm
- Location: Amsterdam, Netherlands
- Contact:
We have successfully performed interference-free EEG recordings with the BioSemi ActiveTwo system near MRI rooms (a distance of 10-20 meters from the shielded room). However, given the large variation in technical details of MRI in setups, I can only recommend to do some actual tests with an ActiveTwo at you specific location. In your particular case, your local representative (www.cortechsolutions.com) will certainly be willing to organize such a test/demonstration.
Best regards, Coen (BioSemi)
Best regards, Coen (BioSemi)
At the moment we are in the process of building new labs for physiological and EGG studies. The Biosemi system with active electrodes look very promising. Can we really use a room without shielding? Of course we can control what is inside the room, but not what is outside.
It is clear that having an amplifier as close to the source as possible can improve the signal quality a lot. I understand that the stray capacitance between the subject and the environment is low and therefor the coupling with low frequency sources is low as well. But what happens when higher frequency fields exist? For example the neighbors are switching on or off an inductive load like an electric motor (airco and cooling systems) of several kilowatts, or fluorescent lightning? This can cause relative strong wide spectrum fields. Can that kind of signals cause interference in the EEG registrations?
It is clear that having an amplifier as close to the source as possible can improve the signal quality a lot. I understand that the stray capacitance between the subject and the environment is low and therefor the coupling with low frequency sources is low as well. But what happens when higher frequency fields exist? For example the neighbors are switching on or off an inductive load like an electric motor (airco and cooling systems) of several kilowatts, or fluorescent lightning? This can cause relative strong wide spectrum fields. Can that kind of signals cause interference in the EEG registrations?