1. What is QEEG brain mapping?

The electroencephalogram (EEG) records human brain electrical activity by electrodes placed over the surface of the scalp, which reflects underlying cortical activity, commonly referred to as “brain waves” (Niedermeyer, 2005). The Quantitative Electroencephalography or QEEG (a method within brain mapping techniques) consists of applying mathematical methods to the EEG data, mainly power spectral analyses, to obtain quantitative metrics that are associated with behavioral-cognitive brain functions. This ultimately may provide electrophysiological markers of disorders such as ADHD, schizophrenia, major depression, traumatic brain injury, and obsessive-compulsive disorder. QEEG is frequently used in biomedical research for measuring the pre and post effects of mental-health interventions. 

The most studied QEEG metrics are the absolute and relative power in different frequency bands (and ratios of two bands). Neuroscientific literature distinguishes delta, theta, alpha and beta frequency bands (Figure 1), and relates them with distinctive states (Niedermeyer, 2005, Kropotov, 2010). For instance:

• Delta band: predominates during deep sleep;
• Theta band: involved in memory encoding and retrieval, and associated with power increases during drowsiness; 
• Alpha band: involved in motor functions (over the motor cortex) and cognitive functioning; 
• Beta band: indicator of inhibitory cortical transmission, and associated with power increases during active concentration.

Figure 1. EEG brain wave patterns for the four frequency bands (delta, theta, alpha, beta). Taken from: https://raphaelvallat.com/bandpower.html

Besides the previous QEEG metrics derived from applying power spectral analysis to the recorded EEG sensors (sensor level), another body of metrics involve the activity in brain structures or areas of the brain (brain source level). This activity can be estimated from EEG recordings using source localization techniques, being LORETA (Low-Resolution Electromagnetic Tomography) a widely used algorithm (Pascual-Marqui, 2002). Consequently, QEEG metrics are represented as two- or three-dimensional brain maps for its interpretation (Figure 2).

Figure 2. Brain maps at sensor level (top) and brain source level (bottom). Top image shows 2D maps representing the absolute power (uV2) in different frequency bands. Bottom image shows the current density power (uV2) in areas of the brain (voxels) for the alpha frequency band, estimated from EEG using LORETA.

2. What is a QEEG normative database?

A QEEG normative database contains a set of metrics computed from EEG and collected from a large number of individuals, large enough to be representative of the population. This allows us to compare an individual’s metric to the QEEG database (also referred to as QEEG testing),  thus indicating whether there exists a non-typical electrophysiological marker with regard to the population. The interest lies in the potential clinical relevance in certain disorders when interpreted by an experienced professional.

A QEEG database is built on a collection of EEG data during active tasks or resting state. 

• An active task refers to an EEG recording of the brain activity while the participant is performing a perceptual, motor or cognitive task.
• During a resting state recording, the participant is in an awake relaxed state with eyes closed (EC) or eyes open (EO). These recordings have the advantage of added simplicity and replicability across laboratories.

The normative term refers to analytical and statistical procedures in the creation of the database to yield valid comparisons (Thatcher, 2003):

• Careful inclusion criteria and representative sampling: Individuals with a history of neurological problems should be excluded, and participant recruitment should be balanced according to demographic variables (gender, age, ethnic background, socioeconomic status, etc). These points are important for the database to generalize to the population.

• Amplifier matching: The frequency characteristics of EEG amplifiers differ (in terms of filters and gain), therefore databases should correct the individual’s QEEG metrics according to these characteristics to make them comparable to the database.

• Approximation to Gaussian distribution: Normative databases perform analytical transformations to gather the QEEG metrics and approximate to a Gaussian distribution (characterized by its mean and standard deviation), which implies a high sensitivity and test-retest reliability (Figure 3).

Figure 3. Distribution of absolute and relative power values (in all frequencies) in the NeuroGuide normative database. Both QEEG metrics have a good fit to a Gaussian distribution. Image taken from Thatcher (2003).

3. How to interpret a QEEG?

Since the QEEG metrics in normative databases follow a Gaussian distribution, an individual can be compared to the population by computing its z-score, i.e., the number of standard deviations that the individual’s metric is away from the mean. Usually, z-scores with an absolute value larger than 2 are considered non-typical (since 95% of values are contained in the [-2, 2] range), with positive values indicating an excess of activity and negative values a deficit (Figure 4).

Figure 4. Plot of a Gaussian distribution, each band has a width of 1 z-score (1 standard deviation away from mean). Based on the figure by Jeremy Kemp, CC BY 2.5. Source. 

As a result, QEEG metrics such as absolute or relative power can be z-scored and represented over two- or three- dimensional brain maps (Figure 5).

Figure 5. Brain maps comparing the absolute power (in different frequency bands) from an individual to the population. Color scale represents z-scores in the range [-3, 3]. Thus, an excess of activity is represented in red color (z-score > 2); a lack of activity in blue color (z-score < -2).

Electrophysiological markers for certain disorders

There is a large volume of research focused on the identification of electrophysiological markers for certain disorders. Recent reviews have pointed out that ADHD in children has shown the most consistent and validated markers across studies, similar to those obtained for schizophrenia; and depression and obsessive-compulsive disorder (OCD) show moderately reliable markers (Newson, 2019; Coburn, 2006).

Below we provide some commonly reported patterns for each disorder. For an extensive review see Newson (2019), Coburn (2006).

• ADHD: The most reliable marker is increased absolute power in slow-wave oscillations (delta and theta) and relative theta power in resting state (eyes closed and open), primarily in frontal regions; and reduced power in beta power in eyes-closed resting state (Barry, 2003). Theta/beta ratio is also considered a reliable marker (Loo, 2012).

• Schizophrenia: The QEEG analysis shows increased absolute power in slow-wave oscillations (delta and theta) and decreased alpha in eyes closed.

• Depression: The dominant marker is increased absolute power in theta and beta for both eyes closed and open conditions, and increased theta power in frontal regions of the brain (using LORETA).

It is important to note that the QEEG is a tool that must be seen as complementary to other clinical observations, and requires the interpretation of the results by a professional.

4. What are the main steps and tools for QEEG?

QEEG analysis requires three main steps:

1. EEG recording: Software should record the electrical activity in the same tasks or conditions that comprise the normative database, recording at least 60 seconds of artifact-free EEG data (Hughes and John, 1999). The EEG amplifier should be matched to the normative database, and, when a brain source analysis is desirable, a high number of sensors distributed over the entire scalp is recommended to obtain a proper estimation (Song, 2016).

2. Artifact rejection: Artifacts are signals recorded along with the EEG but without a neural origin related to the performed task, and are categorized into physiological (cardiac, eye movements, muscle activity) and non-physiological (electric interference, loose electrode contact). These need to be removed before the QEEG testing.

3. QEEG testing and interpretation: The EEG data is processed and artifact-free EEG data is compared to the normative database. The results are then interpreted..

5. Review of QEEG normative databases

Below you can find a short review of three QEEG normative databases currently available in the market. More information can be found in Johnstone (2003), Lorensen (2003); with Keizer (2019) providing a comparison between NeuroGuide and qEEG-Pro.

• NeuroGuide  (Applied Neuroscience, Inc). Contains EC and EO resting-state recordings. Includes 625 individuals (age range from 2 months to 82 years) and provides metrics at sensor level (absolute/relative power and ratios; peak frequency, asymmetry, and coherence) and at the brain source level. Its properties have been reported extensively in the scientific literature (Thatcher, 2003; Thatcher, 2009; Thatcher, 2010) and is FDA registered.

• qEEG-Pro database (qEEG-Pro B.V.). Contains EC and EO resting-state recordings. Includes 1482 and 1232 individuals (eyes closed and open, respectively) in the age range of 6-82. It uses a client-side approach, including new individuals progressively using an automatic artifact filtering. It is FDA registered.

• HBI database (HBImed AG). Contains five active tasks (two GO/NOGO tasks, arithmetic and reading tasks, auditory recognition and auditory oddball tasks) and EC and EO resting-state recordings. Includes 300 children and adolescents (age 7-17), 500 adults (18-60), and 200 seniors (61+).

6. Can I do a QEEG with any existing EEG hardware?

Definitely not. 

Each individual EEG hardware needs to be calibrated to the database to compensate for the measurement differences between the current amplifier and the amplifier that originally collected the EEG data in the database. This process is called amplifier matching, which is always done by the database manufacturer. Once it is completed, the EEG hardware is said to be supported by the QEEG database. 

In this direction, Bitbrain has worked with Applied Neuroscience to include the Versatile EEG 16 and 32 sensors in the QEEG normative database (Neuroguide). Note that the recording with 32 sensors allows obtaining metrics at brain source level using source location techniques (LORETA).

References

• Barry, R. J., Clarke, A. R., & Johnstone, S. J. (2003). A review of electrophysiology in attention-deficit/hyperactivity disorder: I. Qualitative and quantitative electroencephalography. Clinical neurophysiology, 114(2), 171-183.
• Coburn, K.L, Lauterbach, E.C., Boutros, N.N., Black, K.J., Arciniegas, D.B., Coffey, C. E (2006). The value of quantitative electroencephalography in clinical psychiatry: a report by the Committee on Research of the American Neuropsychiatric Association. The Journal of neuropsychiatry and clinical neurosciences 18 (4), 460-500.
• Evans, J. R., & Abarbanel, A. (Eds.). (1999). Introduction to quantitative EEG and neurofeedback. Elsevier.
• Hughes, J. R., & John, E. R. (1999). Conventional and quantitative electroencephalography in psychiatry. The Journal of Neuropsychiatry and Clinical Neurosciences, 11(2), 190-208.
• Johnstone, J., & Gunkelman, J. (2003). Use of databases in QEEG evaluation. Journal of Neurotherapy, 7(3-4), 31-52.
• Keizer, A. W. (2019). Standardization and Personalized Medicine Using Quantitative EEG in Clinical Settings. Clinical EEG and neuroscience, 1550059419874945.
• Kropotov, J. D. (2010). Quantitative EEG, event-related potentials and neurotherapy. Academic Press.
• Loo, S. K., & Makeig, S. (2012). Clinical utility of EEG in attention-deficit/hyperactivity disorder: a research update. Neurotherapeutics, 9(3), 569-587.
• Lorensen, T. D., & Dickson, P. (2003). Quantitative EEG normative databases: A comparative investigation. Journal of Neurotherapy, 7(3-4), 53-68.
• Newson, J. J., & Thiagarajan, T. C. (2019). EEG frequency bands in psychiatric disorders: a review of resting state studies. Frontiers in human neuroscience, 12, 521.
• Niedermeyer, E., & da Silva, F. L. (Eds.). (2005). Electroencephalography: basic principles, clinical applications, and related fields. Lippincott Williams & Wilkins.
• Nuwer, M. (1997). Assessment of digital EEG, quantitative EEG, and EEG brain mapping: report of the American Academy of Neurology and the American Clinical Neurophysiology Society. Neurology, 49(1), 277-292.
• Pascual-Marqui, R. D. (2002). Standardized low-resolution brain electromagnetic tomography (sLORETA): technical details. Methods Find Exp Clin Pharmacol, 24(Suppl D), 5-12.
• Song, J., Davey, C., Poulsen, C., Luu, P., Turovets, S., Anderson, E., ... & Tucker, D. (2015). EEG source localization: sensor density and head surface coverage. Journal of neuroscience methods, 256, 9-21.
• Thatcher, R. W., & Lubar, J. F. (2009). History of the scientific standards of QEEG normative databases. Introd. Quant. EEG Neurofeedback, 2009, 29-59.
• Thatcher, R. W., Walker, R. A., Biver, C. J., North, D. N., & Curtin, R. (2003). Quantitative EEG normative databases: Validation and clinical correlation. Journal of Neurotherapy, 7(3-4), 87-121.
• Thatcher, R. W. (2010). Validity and reliability of quantitative electroencephalography. Journal of Neurotherapy, 14(2), 122-152.