How can the EEG be used in diagnostics?

Through an improved unterstandig of how the brain processes external stimuli and with possibilities of modern technologies, we are at the beginning of an exciting development in which we can use the EEG in the diagnosis.

  • The BEE Group AG...

    … strongly supports the research, validation and use of ERP-based biomarkers in the diagnostics of mental illnesses through investments, especially in the HBImed in Switzerland.

What is the EEG and what is it good for?

Ever Hans Berger discovered and proved the existence of the EEG in the 1920s, it is no surprise that there have been countless hypotheses about what the EEG could be used for. Especially the hypothesis that the EEG could be used for the diagnosis of mental illness has been standing ever since.

However, to date the EEG could only be used to determine three things:

  • whether or not a brain is present,
  • whether or not the brain is working,
  • and to determine whether or not the patient is suffering from epilepsies.

Through modern signal processing and progresses in computer technologies, it is already possible to achieve much more.

Diagnosis using EEG…

High-resolution measuring technology and complex algorithms such as independent component analysis and source localization allows us to determine the observed rhythms in specific areas of the brain. This can be used to observe and analyze the brain’s self-regulating processes. In order to find anomalies in the behavior of a person, however, we require an understanding of how the brain processes external stimuli. This understanding can be obtained through so-called „evoked potentials” or „event related potentials” (ERPs). These are waveforms in a person’s EEG which are triggered (evoked) through sensory perceptions and then correlated with cognitive processes (such as paying attention to a certain task, or the inability to do so).

For this purpose, an EEG is measured in several locations on a patient’s head and recorded while the patient is given a task. These tasks are usually so-called „Go-Nogo“ tasks, i.e. tasks during which the behavior of the patient is assessed based on visual reaction times. In order to be able to retrieve the ERPs, these tasks have to be repeated a large number of times in order to improve the signal-to-noise ratio and average out the spontaneous background EEG. This method has a very high test-retest reliability. Two ERPs measured from a patient under the same conditions will be nearly identical, even if the measurements are taken several years apart.

With this, it is possible to determine the effect of influences on the stimuli processing in the brain, decision-making processes and impulse control to a very high degree of accuracy.

… A new era in psychiatry and neurology?

Now the question is how exactly can ERPs be used to diagnose mental disorder?

The question can be approached with a descriptive example from another field in medicine: A patient has complaints about pains in their arm after an accident. How does the doctor decide whether the patient needs a liniment, a cast or even surgery? In this case, the doctor would take an X-ray photograph of the patients arm to get a better idea of where the pain is coming from. Without the ability to create such a photograph, the doctor would be limited to questioning the patient and trying to feel for any oddities in the arm. The latter roughly corresponds to how mental disorder is diagnosed today. How could we create an „X-ray machine for the brain“, which adds objective measures to the diagnostic process?

Analog to the broken arm example, an approach next to questioning and feeling would be recording the ERPs of the patient who is suffering from a mental disorder and comparing it to that of a healthy individual in the same age group. The ERPs could be used as biomarkers, i.e. as measurable parameters, which are meaningful in terms of prognosis or diagnosis.

The importance of databases in the area of neurodiagnostics

In order to use ERPs for diagnostic purposes, however, there are still two big challenges:

Firstly, databases are required that have a sufficient amount of data from healthy individuals as well as data from people who have been diagnosed with a certain disorder.

Secondly, methods which are capable of recognizing significant differences due to specific disorders have to be devised.

Only when enough data is present can we use it to determine parameters, which are markers for a disorder and support diagnoses with statistically significant readings.

We still stand at the beginning of this development, but the BEE Group strongly supports and invests in the researching, validating and using of these biomarkers, especially through its investments in the Swiss HBImed AG and studies.