Table of contents
1. EEG (electroencephalogram)
1.1 Brain waves
Brainwave (English: brainwave) refers to the electrical vibration generated by the activity of nerve cells in the human brain. Because this kind of swing appears on scientific instruments, it looks like a wave, so it is called brain wave. To describe brain waves in one sentence, it may be said that it is the bioenergy produced by brain cells, or the rhythm of brain cell activity. Every second of human beings, no matter what they are doing or even sleeping, the brain will produce "brain waves" like "current pulses" from time to time. Brain waves can be divided into five categories according to frequency: β wave (consciousness 14-30HZ), α wave (bridge consciousness 8-14HZ), θ wave (subconscious 4-8Hz), delta wave (unconscious below 4Hz) and γ wave (focusing on something above 30HZ), etc. The combination of these consciousnesses forms a person's internal and external behavior, emotion and learning performance.
One downside of EEG devices is spatial resolution—because the electrodes measure electrical activity at the surface of the brain, it's difficult to know with accuracy whether the signal was generated near the surface (in the cortex) or in deeper regions. Of course, some people use calculation methods to invert the collected EEG signals to the location of the brain area. This is possible, but it is not absolutely accurate.
1.2 Artifacts
1.2.1 Eye movement artifacts
Usually appear on the forehead, symmetrically distributed on both sides, and can be identified by wave features.
The methods of exclusion include the following points:
1). The subject is required to eliminate tension and maintain a relaxed state.
2). Ask the subject to lightly press his fingertips on the top of his eyes or eyelids, so that the higher amplitude eyeball waves can be controlled.
3) Place electrodes near the eyes to monitor eye movements and identify the relationship between the eyes and the frontal slow waves.
1.2.2 Myoelectric artifacts
Make the subject relax all over, especially the neck and jaw. Gently close eyes, open mouth, do not swallow, do not clench teeth, etc. Artifacts of a single electrode can be corrected by massaging the skin adjacent to that electrode and replacing the electrode.
1.2.3 Motion Artifacts
The subject can be asked to stop moving, ie motion artifacts can be eliminated.
1.2.4 ECG artifacts
1.2.5 Vascular wave artifacts
1.2.6 50Hz and Static Interference
Installing a shielded room is the most effective method. If EEG is performed in the intensive care unit or operating room, it should be kept away from the power line as much as possible to reduce the swing of the electrode line and the charge movement of the liquid or other objects in the container.
1.3 Artifact removal method
At present, the methods for dealing with artifacts mainly include: avoiding artifact generation, direct removal method, and artifact elimination method.
1.3.1 Avoiding Artifacts
Artifact avoidance refers to telling subjects not to blink or do some actions that may produce artifacts before the experiment. This method plays a certain role in avoiding artifacts, but some artifacts such as ECG are unavoidable.
1.3.2 Direct removal method
The direct removal method (artifact rejection) refers to finding the EEG signal segments containing artifacts through observation or automatic recognition, and directly deleting these signals.
The direct removal method can only remove some obvious signals containing artifacts. Moreover, this method also has a defect, that is, this method will discard a large number of useful signals, which is not conducive to EEG experiments with relatively few signal data.
1.3.3 Artifact removal method
Artifact removal refers to directly identifying and separating artifact signals from EEG signals. It separates artifact signals while retaining useful EEG signals. It is an ideal method for removing artifacts.
1) Regression method: The regression method based on time domain or frequency domain can remove the artifacts related to eye movement, but the disadvantage of this method is that it will delete the EEG signal related to electrooculopathy while removing the artifacts of electrooculogram.
2) Threshold method: Threshold method can also eliminate interference such as oculoelectricity. The basic idea is to discard the recording time period whose amplitude exceeds the normal range. The disadvantage is that the information of the corresponding time period is also lost.
3) Time-domain signal processing method: Time-domain signal processing includes principal component analysis (PCA) and independent component analysis (ICA). Their core is to decompose EEG signals and artifacts into different signal components, and then eliminate them. The classic ICA algorithms are: FastICA algorithm, JADE algorithm, extended maximum entropy algorithm and informax algorithm.
2. SEEG (stereoscopic electroencephalogram)
Stereo-electroencephalography (SEEG) is a minimally invasive surgical procedure used to identify the brain region where seizures originate. During SEEG, doctors place electrodes on targeted brain regions, which are then monitored to pinpoint the source of seizures. SEEG can find seizure sites deep in the brain that conventional electroencephalography (EEG) tests may not be able to reach. It covers both sides (hemispheres) of the brain. To implant the electrodes, the surgeon makes 10 to 20 (depending on the patient) small incisions in the scalp and skull to minimize blood loss. Electrodes are placed through small holes, allowing the exploration of large areas of the brain with minimal tissue damage.
3. CT (computed tomography) scan
Medical professionals use computed tomography (also called a CT scan) to examine the structures in your body. A CT scan uses X-rays and a computer to generate images of cross-sections of the body. The pictures it takes show very thin "slices" of your bones, muscles, organs, and blood vessels so healthcare providers can learn more about your body. Traditional x-ray machines use a fixed tube to point x-rays at a point. As X-rays pass through the body, they are absorbed in different amounts by different tissues. Against the black background of film, denser tissues produce whiter images than other tissues. X-rays produce two-dimensional images. A CT scan has a donut-shaped tube that rotates the X-rays 360 degrees around you. The captured data provides a detailed 3D view of the inside of your body. Sometimes your scan uses contrast material. This contrast agent, sometimes called a dye, improves images by accentuating certain features. Depending on the type of CT scan and the reason for the scan, your healthcare provider will give you a special fluid that contains contrast material to drink, give you a contrast medium intravenously, or both. The contrast material is eliminated from your body through urine, first quickly and then more slowly over the next 24 hours.
4. MRI (Nuclear Magnetic Resonance)
Magnetic resonance imaging (MRI) is a new inspection technology based on the principle that atomic nuclei with a magnetic moment can produce transitions between energy levels under the action of a magnetic field, and use the interaction between an external magnetic field and an object to image. Functional magnetic resonance imaging, or functional MRI (fMRI), measures brain activity by detecting changes associated with blood flow, a technique that relies on the combined fact that cerebral blood flow and neuronal activation increase when an area of the brain is in use.
MRI is an imaging technique that uses a strong external magnetic field to pulse hydrogen protons to produce magnetic resonance phenomena. CT irradiation is accompanied by ionizing radiation, which will cause certain radiation damage, while no research on the magnetic field of MRI has clearly shown that it will cause damage to the human body.
Due to the different imaging principles, CT and MRI are very different.
1. Application method: There are two clinical application methods of CT, namely plain scan and enhanced scan. There are various scanning methods of MRI, including ordinary plain scan, fat suppression examination, water suppression examination, susceptibility weighted imaging, contrast-enhanced examination, MRA examination, etc.;
2. Resolution: CT images have higher density resolution, and CT has a higher advantage in density resolution. MRI images have high soft tissue resolution, and are significantly better than CT in displaying central nervous system and intra-articular structures and lesions;
3. Image orientation: CT mainly uses axial tomographic images, while MRI can directly obtain tomographic images in any orientation, which is conducive to displaying the anatomical relationship between tissues and structures, and is also conducive to clarifying the origin and scope of lesions;
4. Parameter selection: only density, window width and window level parameters can be adjusted for CT inspection. MRI examination has multiple imaging parameters, and each parameter provides different information. Combining images with different parameters is helpful for the detection, diagnosis and differentiation of diseases.
In addition, CT and MRI have their own characteristics. For example, the density of CT images can be quantitatively evaluated. Using the density quantification unit of CT values can clarify the nature of different densities and distinguish soft tissues with small density differences from their lesions. MRI does not have this function. On the other hand, MRI can directly perform water imaging. Using T2WI sequence, the pipeline system containing fluid can be displayed as a whole without any contrast agent. This is MRI water imaging, which is conducive to observing the bile duct, urinary tract, and spinal cord.
How does an MRI work?
MRI is a complex imaging method, here we give a brief overview.
As the name suggests, the magnets are at the heart of MRIs, but they are much stronger -- about 1,000 to 3,000 times stronger than ordinary refrigerator magnets. The magnetic field produced by the MRI interacts with the protons in the hydrogen atoms (and our bodies are about 70% water, which is very useful - magnets affect a lot of hydrogen atoms).
Usually, these protons face randomly, but when a magnetic field is added, a large number of them align in the same direction under the action of the magnetic field. So, when we're lying in an MRI machine, the protons in the hydrogen atoms (which are in the water inside us) point in roughly the same direction.
And then, a radio pulse is emitted (just like a normal radio signal, but faster), which also interacts with the protons, essentially turning them sideways. However, because the radio frequency occurs only for a moment, the protons relax back to their previous alignment.
As the protons relax, energy is released, which is detected by sensors in the MRI machine. With a few calculations, the computer can determine what the tissue looks like based on the energy released, and show us a picture of the tissue.
fMRI (Functional Magnetic Resonance Imaging)
If a person wants to move his right arm, something needs to happen: Some part of his brain will increase its activity, sending messages to complete the movement, and that area of the brain will receive slightly more oxygen-rich blood than before.
With fMRI, the same thing happens with MRI -- the energy released by the relaxation of protons is measured -- but the purpose of the calculation is to determine how oxygenated blood flow changes.
If one part of the brain has more oxygenated blood than other parts, that area of the brain may be more active. This is called blood oxygen level dependent (also known as Bloodoxygen level dependent, BOLD).
This is what we see in fMRI, which is usually seen on MRI images.
One disadvantage of fMRI is temporal resolution. Data collection is slower because changes in blood flow take several seconds and the actual recording is limited by computational factors.
This typically means that participants are stimulated multiple times, and their brain responses are recorded at different time points each time (e.g., recording responses at the onset of the first stimulus, 10 ms after the onset of the second stimulus, etc.).
This may undermine the accuracy of recording new responses, but it does provide a full range of brain responses.
How do they compare?
As we've covered above, there are several differences in the way each technique provides brain imaging information.
There are other things to consider (like cost) - MRI machines cost much more (both to buy and to maintain) than EEG (electroencephalography) machines, and require a much higher level of training.
Field work with MRI/fMRI won't happen either, because they aren't portable enough.
Experimenting with an EEG doesn't require much trouble, either—sometimes it's just a matter of putting on a headset and checking the quality of the data. Automatically calculated metrics can also provide quick insights into human behavior through EEG.
5. fNIRS (functional near-infrared spectroscopy)
fNIRS refers to the use of near-infrared spectroscopy (NIRS) for functional neuroimaging purposes. Using fNIRS, brain activity is measured through hemodynamic responses that correlate with neuronal behavior. It is a neuroimaging method that uses spectroscopy to measure the level of neural activity in the brain.
fNIRS is sensitive to hemodynamic responses to brain activation. The technique also has the ability to distinguish between changes in oxyhemoglobin and deoxygenated hemoglobin. This device has been successfully implemented as a control signal for the BCI system.
Functional near-infrared spectroscopy is an emerging non-invasive brain functional imaging technique in recent years. The principle of fNIRS for brain functional imaging is similar to that of fMRI, that is, brain neural activity will lead to local hemodynamic changes. It mainly uses the difference in the absorption rate of oxyhemoglobin and deoxygenated hemoglobin in brain tissue to near-infrared light with different wavelengths of 600-900nm to directly detect the hemodynamic activity of the cerebral cortex in real time. By observing this hemodynamic change, that is, the neural activity of the brain can be deduced through the law of neurovascular coupling.
For example, when a subject is asked to perform a right-hand finger movement task, the left side of the cerebral cortex discharges motor activity, consuming oxygen and energy. At this time, the overcompensation mechanism of the blood supply system in the brain will input a large amount of blood rich in oxyhemoglobin to the local area, resulting in an increase in the concentration of oxyhemoglobin in the area and a decrease in deoxygenated hemoglobin. In the fNIRS experiment, the experimenter asked the subjects to perform tasks according to a certain experimental paradigm, and at the same time used fNIRS to observe the changes in the concentration of hemoglobin in different parts of the brain. If a certain brain region is found, its hemodynamic activity is highly related to the task design, and it can be inferred that this brain region is activated by the experimental task.
How does fNIRS work?
The main physiological parameters it detects: the concentration changes of absorbing chromophores (such as HbO2, Hb, totalHb, etc.) in tissues.
Measurements of local blood flow signals are performed by light sources and detectors placed on the head of our brain.
This usually means measuring the size of the brain area, depending on the arrangement of light sources and detectors and the support of the equipment. The biggest advantage of fNIRS among these technologies is that its time resolution is faster than that of fMRI technology, its spatial resolution is larger than that of EEG technology, and more importantly, its portability and artifact interference are small.