QEEG guided neurofeedback training
QEEG (quantitative electroencephalography) can be used in a number of ways including to identify a neurofeedback strategy. This article shows how the NeXus-32, Mind Media's premier professional unit can be used in a practical scanario with ADHD.
The NeXus 32 has 24 unipolar channels to support a "full cap" EEG recording using the 10-20 protocol for electrode placement.
The NeXus Cap Adaptor allows easy connection of the EEG cap to the NeXus 32. Suitable steps should be taken to prepare the skin and the electrodes which are not described here. (See accessories for suitable materials)
Of course, that still leaves some channels that can be used for other physiological signals to be recorded simultaneously along with the EEG if necessary. These could be deployed as follows:-
- Four EXG inputs that can support EMG, EKG, EOG etc.
- Four Auxiliary inputs that could support Skin Temperature, EDA, Respiration Rate or BVP measurement
Many clients can benefit from some relaxation training prior to commencing any neurofeedback work and the fact that this range of signals is available means that a stress test protocol, for example, can be used.
QEEG versus EEG
The EEG mesurements are made using what is called a "montage" - which means a rule according to which EEG potentials at the electrode sites are measured. The simplest and often used montage is a "Linked Ear" approach. In this montage the electrode potentials at individual electrode sites are measured with reference to two linked electrodes located at the left and right earlobes.
In a typical situation the raw EEG is collected by the NeXus 32 under control of the BioTrace+ software for a few minutes. If you have ever looked at raw EEG you will realise it is not a trivial matter to identify patterns of significant change in that data. It is also necessary to identify and eliminate or at least minimise artefacts due to issues such as eye movement, muscle contractions etc that can contaminate the raw data.
In order to Quantify EEG, the first step is usually to derive some form of frequency spectrum via a Fourier Series analysis. This may represent the individual signals as broken down into the relative power in appropriate frequency bands. A second step could be to compare the captured data to one of the existing QEEG databases such as Neuroguide.
When EEG is recorded from many electrodes across the whole cortex it is possible to compute a 2D map of the measured EEG characteristics. The map could for example display a potential value or averaged power. However, recording EEG and frequency spectra are often only the initial stages of QEEG analysis. The most important stage s comparing the data collected from a subject with selected "normative data".
The concept and practical implementation of a normative database is challenging. It raises the the issue of what we mean by "normal". it is assumed that a suitable database consists of data from many individuals in a population without brain damage and screened for neuropsychiatric disorders etc. The database must represent a suitable mix of ages, genders, ethnicities and other demographic factors. A particular challenge is the paediatric population where significant developmental changes occur over a few years.
If you imagine we have a suitable database, any parameter may then be considered to fit a normal, Gaussian distribution. If for an individual parameter, for example, the EEG power in a particular frequency band at an electrode site, we divide the fluctuation from the mean by the standard deviation, the result is a measure of fluctuation from normality for this individual electrode site. This is called a Z score and we are interested in whether the particular measures we have taken are significantly different from what we would expect of the statistic norm.
There are a number of normative databases available and one commonly used with the NeXus 32 is the Neuroguide database developed by Robert W Thatcher. This is commercially available and includes data from over 600 individuals covering an age range of 2 months to over 80 years.
A ten year old female diagnosed with ADHD exhibits some issues with symptoms as follows:-
- Easily distracted
- Restless, moves around a lot
- "Dreamy", stares into space, finds it hard to maintain attention
- school problems with maths, reading, telling the time, days of the week.
- seems to need to time to process information coming from her environment.
Using a Z Score approach and examining the power in various frequncy bands we end up with the 2D maps shown above. The maps show high power (Bright red) in the parietal cortex compared with the norms in the frquency range 20 to 30 Hz (High Beta). This is also supported by neuropsychology data collected.
This approach suggested a particular protocol to inhibit frequencies int he 20 to 30 Hz range using electrode locations P3 and P4.
The training then involved watching a movie with a feedback threshold set to allow the movie to play whilst the High Beta energy stays below the threshold and pause when the frequencies exceed the threshold. The protocol was followed twice per week for 40 sessions.
The EEG is more normalised following training with a reduction of the High Beta around the parietal cortex. Of course what actually matters is not the change in EEG patterns but any change in symptoms - in this case there was a significant reduction of reported problems.
Parents and child are very enthusiastic about the results of the training
- Concentration is improved, better able to keep her attention
- Less daydreaming
- Better school results, especially math
- Child is in a better mood
- Child makes better social contact, seems to got out of her ‘own world’ and makes a better connection with the world around her
If you would like to know more about a comprehensive system for QEEG and biofeedback plus neurofeedback you can review this link - Comprehensive qEEG plus Neuro and Biofeedback Package and request a quote.