Fullscreen HTML5Med Res (??MB)HTML4

How to monitor your brain waves An Introduction to Electroencephalography (EEG) By Jim Rowe Elsewhere in this issue, we describe a low-cost Brainwave Monitor which you can build to measure and record brainwaves – yours, or those of someone else. But to use that device, you need to understand what an EEG is, how to use it and how to interpret the results. This article explains what an EEG is all about. E lectroencephalography or on the exposed brains of the animals. “EEG” involves monitoring the Beck is generally credited with proposelectro-neurological activity of ing the concept of brain waves. German physiologist and psychiathe brain, using electrodes placed in trist Hans Berger was the first to record strategic positions on the scalp. This is not to be confused with the human EEG signals in 1924 and also ECG, or electrocardiograph, which the first to coin the term “electroencephalogram” to describe the function monitors the tiny electrical signals of the machine he developed. which control the heart. His recordings were made using But an absence of either EEG or ECG electrodes placed on the subject’s signals means you’re just as dead! The first person known to try look- scalp, rather than on the surface of ing for electrical activity in brains their exposed brain – a far less invasive was British physician Richard Ca- scheme, making it much more suitable ton, who did experiments on the ex- for use on human subjects! Since Berger’s pioneering work posed brains of rabbits and monkeys in 1875. He published his results in there has been a lot of development the British Medical Journal in August, of EEG measurement technology and 1875 (siliconchip.com.au/link/ aakh). Then in 1890 Adolf Beck, a Polish physiologist, published the results of tests measuring electrical activity in the brains of rabbits and dogs – including rhythmic activity altered by light striking the animals’ Fig.1: tiny signals within the eyes. As with Caton’s work, this brain are passed from axon to dendrite. was done by placing electrodes These are detected and read as an EEG. 14 Silicon Chip Australia’s electronics magazine the application of EEG recordings for diagnosing various neurological and mental health problems. Nowadays, it is used for such diverse things as distinguishing between epileptic and other types of seizures and in the analysis of sleep disorders. The American EEG Society was founded in 1947 and the first International EEG congress was held in the same year. There are now EEG Societies in a number of countries, as well as internationally recognised techniques regarding the placement of EEG electrodes (described below). How an EEG works Our brains are made up from billions of nerve cells or “neurons”, which constantly communicate with one another by transferring ions between them via the tiny gaps or “synapses” separating them. At one end of the synapse gap is a tentacle-like axon (a protrusion of the neuron cell) while the receiving site on the siliconchip.com.au other side of the synapse is known as a dendrite. Since the ions are electrically charged, this means that there are small electric currents flowing all the time – especially in the outer layers of the cerebral cortex, which is the outer ‘grey matter’ part of the cerebrum (the large upper part of the brain). Although these currents are quite small, a proportion of them passes through the meningeal envelope surrounding the brain and out through the bones of the skull cap and the skin of the scalp. As a result, minute voltages corresponding to these currents can be detected using electrodes attached to the scalp, as shown in Fig.2. Because these voltages are so tiny, a great deal of amplification is needed to sample and record them. This means that it’s essential to use various techniques to cancel out “common mode” signals, such as voltages induced by nearby 50/60Hz mains wiring, which would otherwise drown out the EEG signals. The frequencies of the EEG signals are quite low, varying between about 0.5Hz and 16Hz. This means that lowpass filtering can also be used to reject 50 or 60Hz hum. So the basic idea of EEG is to monitor brain activity by using an array of small electrodes placed on the subject’s scalp, to sense the leakage voltages present on the surface. Electrode placement You cannot just stick the electrodes Fig.2: electrodes placed on the scalp are used to detect tiny voltages caused by currents flowing between neurons in the outer layers of the brain’s cerebral cortex. A small fraction of these currents passes out through the meninges, the skull cap and the scalp. anywhere on the scalp. You must follow the standardised placement of EEG electrodes on a patient’s scalp, to allow comparisons and diagnoses to be made. The most common EEG electrode placement standard used nowadays is called the International 10-20 System, which is as follows. Fig.3 shows two views of a stylised human head, from the side and from above. Three main reference points are shown: the “nasion”, the “inion” and the “vertex”. The nasion is the depression directly between the eyes, just above the bridge of the nose. It’s the intersection of the frontal bone and two nasal bones and is regarded for EEG purposes as the landmark for the front-centre of the skull. The inion is the location of a small bump or protuberance on the outer surface of the occipital bone of the skull, which can be felt through the scalp. This point is regarded for EEG purposes as the rear centre point of the skull. The vertex or top centre of the skull is basically the point halfway along the centre line of the skull, equally distant from the nasion and the inion. This vertex is used to locate the reference ground (Cz electrode) for EEG Fig.3: EEG electrodes should be placed on the scalp in positions defined by the International 10-20 System, and illustrated here. siliconchip.com.au Australia’s electronics magazine August 2018  15 Fig.4: the combinations of EEG electrode positions which are most useful for sensing slow waves, ‘spindles’ and Alpha rhythms. Note that the Cz ‘reference ground’ electrode should always be placed at the skull’s vertex. measurements. This is used as the basis of the 10-20 EEG electrode placement grid. The distance between the vertex and the nasion is divided into three parts, with intervals of 20%, 20% and 10% as shown, and in the same proportions for the distance between the vertex and the inion. Similarly, the distance between the vertex and the line on each side of the head between the nasion and inion is also divided into three parts with intervals of 20%, 20% and 10% as shown in Fig.3. These points are then used to visualise a grid, as indicated by the dashed red lines on each view. The intersections of these grid lines are used for most of the EEG electrode positions. These are labelled using a convention where electrodes on the longitudinal centre line have the suffix “z” (as in Fz, Cz and Pz), while those on the left-hand side of the skull are given odd numbers (like F3, C3, P3, F7, T3 and T5) and those on the right-hand side are given even numbers (like F4, C4, P4 and so on). The letter prefixes given to these electrode positions correspond to the names of the brain lobes underneath their positions. So the electrodes above the frontal lobes are given the prefix “F”, those above the temporal lobes have the prefix “T”, those above the parietal lobes have the prefix “P” and those above the occipital lobes have the prefix “O”. In addition to the 19 electrode positions defined by the 10-20 grid, there are four extras; two near each ear. As shown in Fig.3, these are M1 and M2, located at the left and right mastoid protuberances (the small bumps just behind and above each external ear), and A1 and A2, located either on the lobe of each external ear or on the tragus, the small pointed skin protuberance just above and behind the lobe. In practice, the M1 and A1 electrode positions are regarded as interchangeable, as are the M2 and A2 positions. This is because they are both very near the midpoint of the lowest grid line between the nasion and inion on each side, ie, two near each ear. Note that for higher-resolution EEG measurements and research, many additional electrode positions are used. Generally, these are located halfway between the grid lines shown in Fig.3. The additional electrode locations are labelled according to the Modified Combinatorial Nomenclature (MCN). But this more complex electrode array system needn’t worry us here. Which combinations are useful? With so many electrode locations to choose from even in the 10-20 system, selecting the combinations which are likely to be the most useful can be a bit bewildering. Fortunately, people who have recorded a lot of EEGs over the years have come up with a short list of electrode combinations that have been found most useful. These are listed in Fig.4 – Suggested Electrode Combinations. The combination of F4 and M1 (or A1) is suggested as best for capturing slow EEG waves, with the F3 and M2/ A2 combination as an alternative. Similarly, the combination of C4 and M1 is suggested as best for capturing rapid “spindle” EEG waves, with the C3 and M2 combination as an alternative. Then for capturing the brain’s relaxed “alpha rhythm”, the combination of either O2 and M1 or O1 and M2 is suggested. With all of these combinations, the EEG sampler’s ground reference lead is assumed to be connected to the Cz electrode at the vertex or top of the skull. This is necessary to achieve the clearest and least noisy recordings. So you don’t need a huge number of electrodes and leads to capture the most useful EEG recordings. In fact, with only seven electrodes (including the Cz electrode), you can perform three different EEG recordings simultaneously, using an EEG Sampler with three differential input channels. Stimulating neurons electrically While this article is about sensing the electrical impulses generated by neurons, it is also possible to do the reverse, ie, use externally-generated electrical impulses to stimulate neurons. We described a circuit to do just this in the project about Cranial Electrical Stimulation (CES) in the January 2011 issue (siliconchip.com.au/Article/871). This is intended to reduce the pain from headaches and to promote relaxation. In addition to the synapses described earlier, for communication between neurons, synapses also exist between motor neurons and muscle fibres. The electrical impulse across the synapse causes the muscle fibre to contract and this is how the brain controls movement in the body. The injection of an electrical im16 Silicon Chip pulse along this path can cause the muscle to contract involuntarily. Similarly, sensations such as heat, cold and pain cause electrical impulses which travel to neurons in the brain via synapses. We have previously published two circuits for transient electrical nerve stimulation (TENS), which can be used for pain relief. See the August 1997 (siliconchip.com.au/Article/4848) and January 2006 (siliconchip.com.au/Article/2532) issue for details. A warning: as you will note in the TENS articles, their output must NEVER be applied to the head, especially in the areas where EEG electrodes would go. NEVER try to connect a TENS machine to EEG electrodes (in most cases, they won’t fit anyway!). Australia’s electronics magazine siliconchip.com.au But why would YOU bother? While it should be pretty obvious that an EEG in the hands of a medical professional would be extremely valuable in all sorts of clinical/diagnostic situations, the question must be asked, “why would the average person bother reading their (or someone else’s) EEG?” And “don’t you need many years of experience to decipher EEG waveforms?” In a professional application the answer to the latter question is undoubtedly yes – it would be folly (and probably dangerous!) for an untrained person to even attempt to analyse EEG waveforms with a view to diagnosing brain disorders. However . . . Fig.5: sample waveforms showing how EEG waves change during the various stages of relaxation and sleep. Our Brainwave Monitor is designed for this exact task. It’s possible to switch each of the Monitor’s three input channels between two alternative electrode pairs, using a small electrode switch box to be described in a future issue. Then by using only three additional electrodes and leads (ten in all), you can capture EEGs from any of the electrode combinations shown in Fig.4, merely by selecting them using the switch box. What to look for So what kind of EEG waveforms can you expect when using the Brainwave Monitor? We can’t explain everything you need to know to interpret EEG waveforms in this article – that’s a job for an expert. But the waveform samples shown in Fig.5 will give you an idea of the sort of waveforms you are likely to see at various stages of brain relaxation and sleep. EEG waves are named according to their frequency range. They are Delta waves if their frequency is between 0.1Hz and 3.5Hz, Theta waves if their frequency is between 4Hz and 7.5Hz, Alpha waves for frequencies between 8 and 13Hz and Beta waves in the range 14-40Hz. Their peak amplitude is typically between 10µV and 100µV, with Alpha waves generally less than 60µV and Beta waves usually in the range 10-20µV. So an amplification factor of around 5,000 to 250,000 times is required for the EEG signals to be sampled by a typical analog-to-digital converter (ADC). As you would expect, the signal amplitudes are greater if measured at the surface of the brain (1-2mV). Even this is a small fraction of the voltage of a nerve impulse, which is around 100mV. In spite of the problems of amplifying and processing such tiny signals in a very noisy electrical environment, our Brainwave Monitor makes this a reasonably routine procedure. You can connect it to your laptop or notebook PC to view and record brainwave signals. What a great idea for a school electronics project! siliconchip.com.au There are many references (on the net and elsewhere) extolling the virtues of a personal EEG in controlling and changing your own brain activity. Possibly using external simulation, with practice it appears you can “train” your brain to achieve some positive outcomes. Indeed, there are several commercial organisations which offer various EEG-compatible software to enable users to experiment in this area – the example below is from the US Transparent Corporation (www.transparentcorp.com) who claim that EEG units can be tools to improve the mind through a non-invasive brain stimulation process. “Neural stimulation therapy, also commonly referred to as brainwave entrainment, uses deliberately engineered sound or light stimuli to influence the mind in beneficial ways”. Other reports we’ve seen suggest EEG can be used for highly stressed individuals to reduce those stress levels by recognising the types of EEG waveforms which not only reveal stress but also the waveforms which show stress reductions. We’ve also seen claims that EEG analysis can help those suffering sleep disorders. There are also reports of students who use EEG to reduce stress levels before important exams. And others which show that a general sense of wellbeing can be achieved by knowing what brainwaves show. We’re not saying that these reports are all accurate (indeed, any of them!) – the net is notorious for misinformation – but if you’re interested in these, or many other “self-help” applications of the EEG, we would strongly suggest you do extensive study so that you know what you are doing. It might also be wise to discuss any possible plan of action with a health care profesSC sional who has expertise in this area. Using Transparent Corp’s “Emotiv EPOC or Emotiv EEG for EEG-Driven Stimulation” Australia’s electronics magazine August 2018  17