There are a bunch of techniques (that I know of), and none of them are exactly easy to explain. Let me start by listing the types I know of – they can easily be found on the internet :
Now, for the details. I will do my best to break each down into as basic a description as I can, but I may overlook some concepts. Let me know. I’m not going to go into much detail on the diagnostic uses for each technique, but I will make a mention of at least one possibility for each.
1) EEG – Electroencepholography
Neurons within the brain use voltage potentials to transmit signals. As some ions enter the neuron, other leave, causing a voltage change. Such a flux of charged ions causes local changes in voltage. These voltage changes, small as they may be, radiate out from the neurons themselves to create a local change in voltage. As such, the voltage changes can be picked up by specialized electrodes which are sensitive to small changes in voltage.
An EEG is usually achieved by placing a cap covered with highly sensitive electrodes on the shaved head of the patient – 1) so that hair does not interfere and block signal transmission, and 2) so that a conductive gel primer can be applied, reducing the capacitance of the skin (when applicable to the study). Depending on the number of electrodes attached to the cap, which basically places an upper limit on the resolution of your measurement, you can measure the constant fluxes of activity in the brain. It’s hard to pinpoint the exact origin of the voltage fluxes because the cap does not encapsulate the entire head; just the top of the scalp. Also, an EEG cannot tell you much about regions deep within the brain unless those regions have large groups of neurons all activating in synchronous action. Since the change in ions would diminish over distance, it would take a very large flux of ions from the center of the brain to cause a voltage change on the surface. Still, that’s why very sensitive leads are used.
However, certain groups of neurons cause distinctive voltage fluctuations when activated. This is how neuroscientists can parse the different voltages. For example, different aggregated brain waves (i.e. Delta waves, Beta waves, etc) operate at different frequencies. An EEG can parse out these frequencies, assuming the signal is passed through the proper filter (ideally a bandpass filter that can eliminate residual artifacts as well as removing any movement by the subject). At any rate, an example EEG output may look something like this:
It should be clear from the above image that there are 1) differences between sleeping and awake states (the first half versus the second half of the graph), and 2) differences between the frequencies of the signals produced by each particular brain wave. Based on brain wave functionality, researchers assume correlations between external stimuli and brain responses. It’s not perfect, but can be useful.
2) CAT – Computerized Axial Tomography
Basically, a CAT scan is a scan created by using simple 2D x-rays of the brain and then using computer software to align all the images to create a 3D image. The patient lies still in a donut-shaped x-ray machine and the machine takes successive images of the patient’s brain. As it moves in a circle around the patient, a computer program must be running to keep straight where the x-ray is positioned at any given time so that the reconstructed image is accurate. I don’t know how many x-ray images are actually taken, but the machine must go through at least one complete rotation to ensure it captures all of the brain pieces it can.
My guess would be that the amount of images actually taken depends on the resolution the doctor would want. Because the actual x-ray emitter takes up space, it is not possible for the machine to take, say, 360 degrees worth of images in one pass. A reasonable assumption would be that the camera is offset by about 20-30 degrees each time it completes a loop, so depending on how accurate the scan needs to be, you’re looking at anywhere from 3+ loops – say the scanner takes up 20-30 degrees of the circle, depending on the size of the machine, then each rotation of the scanner would take 13-12 images.
Once the imager (the actual x-ray machine) is done, the computer can complete its work and reconstruct all the 2D images into a 3D display. The end result, depending on the computer software used, is either a plain black-and-white reconstruction of the brain (useful when looking for ossification or any major deformations) or a more colorful reconstruction (useful for monitoring regional blood flow changes):
3) PET – Positron Emission Tomography
Similar to a CAT or CT scan, a PET scan sometimes uses x-ray images in a 360 degree coverage circumference to image the brain as a whole (of course, needing a computer to compile all the images to create a 3D model). The difference here is that PET scans rely on gamma radiation released from radioactive isotopes injected into the patient.
Of course, the tracer isotopes injected are not particularly dangerous (or the technique would be useless), but they serve an important purpose: they tell the doctor what parts of the brain is being used at a given time. For instance, if a patient is injected with radioactive glucose – the body’s primary source of energy – then any area of the brain that is currently active would emit gamma rays. Depending on the amount of radiation being collected from specific areas of the brain, the doctor can tell which regions of the brain are more active at a given time. -> In order to function, the neurons in the brain need glucose. The more active the neurons are, the more glucose they take in and break down for energy. The more glucose a cell or group of cells takes in, the more radiation is emmitted from that particular region of the brain.
It is in this manner that a doctor, after the images are compiled and reconstructed, can tell which areas of the brain are more active than others – the active regions will ‘glow’ or show more radiation emissions.
Of course, an x-ray machine alone cannot detect this gamma radiation, so on top of the donut-shaped x-ray machine, a special radiation detector must be integrated into the circular design of the standard CT imaging machine. These radiation detectors are integral – as they tell the computer when the radiation is being collected. By matching the timing of the radiation with the images taken by the CT scanner, the computing software can reconstruct a 3D image that shows regions of increased metabolism in the brain for any given time lapse. Typically, regions of high activity are represented by red, whereas decreased activity shifts into the blue shades:
4) MRI – Magnetic Resonance Imaging
Perhaps my favorite, the MRI scans use a 2 step system to collect their images. The only constraint is that the patient be laying flat and keep his head stable throughout the entirety of the scans. That said, there are two key concepts behind the MRI.
First, as implied by the name, once the patient is in the machine, and MRI creates a strong magnetic field longitudinally (in line with the patient’s body). The field is strong enough to align the majority of the water molecules in the patients brain to be in line with the magnetic field*. However, not all of the water molecules end up staying in line with the field; there are just so many molecules that it stands to reason that there would be a significant amount that, for whatever reason, do not align with the magnetic field.
Second, the MRI sends a potent radio signal into the patient at the level being scanned, which causes those particular molecules that are not in alignment to energize and spin haphazardly. When the radio signal is stopped, the spinning molecules calm down. In order to calm down, these molecules need to lose energy, so they release radiation (I think it’s usually photons, as electrons that drop energy states typically shoot off protons). The MRI picks up on this radiation and uses it to construct an image at that particular region of the brain.
This process is then repeated, starting at the tip of the head, and typically moving down into the neck to ensure the entire brain is scanned (as far as I know). As with all the other imaging techniques, an MRI creates a series of 2D images. However, these images are in order, so the computer program that reconstructs them to create a 3D interpretations doesn’t have as hard a job as in other scanning techniques. In the end, the doctor gets a series of 2D images he can scroll through to see if there are any abnormal growths (which appear as specks, then grow as you progress down the patient’s body, then fade away as you scroll past the spatial regions in which the growths are located). There is also a 3D image the doctor can look at; though most of my work, and the work of radiologists, consists of scrolling through the 2D images.
The images created and captured by an MRI are surprisingly detailed, showing different levels of opaqueness depending on the tissue of the region (i.e. bone is brighter than nerve cells, large aggregations of nerve cells are brighter than less populated regions, etc). Though it may be in black and white, the image created by an MRI is incredibly clear in its contrast, allowing for a lot of minuscule objects/anomalies to be seen.
* Water molecules are characterized by partial charges (delta +/-). The oxygen molecule, being larger and spatially distant from the hydrogen atoms, is more negative, whereas the hydrogen atoms, being distanced from the oxygen, are slightly positive:
A strong magnetic field, such as that in a common microwave, can manipulate water molecules based on the partial charges.
5) fMRI – function Magnetic Resonance Imaging
Similar to an MRI, an fMRI uses the innate magnetic properties of hemoglobin (the proteins within blood that transport oxygen) to measure blood flow in regions of the brain. Each hemoglobin molecule contains an iron molecule, which is ferromagnetic, so it responds readily to an induced magnetic field.
Rather than worry too much about picking out individual region of the brain, an fMRI is more concerned with regional brain activity. In this case, brain activity is assumed to be directly correlated with blood flow through the region: the more blood flow through a specific region, the more active that region is, compared to other areas. Ostensibly, this makes sense; just as PET scans measure the metabolic activity of the brain, they glucose needs oxygen to be broken down, so an increase in blood to that region must accompany the glucose or function would remain at baseline.
What’s important to note is that, in general, since an fMRI is concerned with regional activity and not acute spatial recognition and separation, detail is often sacrificed for the sake of capturing as much response to hemoglobin as possible. See the following image and compare it to that of a standard MRI above:
Again, as in the PET scans, the brighter, more red/orange areas indicate higher levels of activity whereas the darker, blue areas indicate less/little activity
6) MEG – Magnetoencephalography
A newer development (of which I have little knowledge), the MEG picks up on the weak magnetic fields that are created by local voltage changes in the brain. As background: a change in voltage induces a change in the local magnetic field. As such, if a group of neurons activates at once, then it send a small magnetic field ripple, emminating from the center of these neurons. The MEG uses highly sensitive magnets to pick up on this shift in the magnetic field and tries to triangulate the position of the location of the magnetic shift.
As far as I know, the technology behind this is still relatively new, but the idea is simple: create a machine that picks up on the small magnetic changes that occur when a neuron fires. I’m guess that, at this point, the MEG is only able to pick up very large groups of neurons firing at once. Anything else would likely cause too small a disturbance and be disregarded as noise.
2.ERP (eventrelated potentials)
4.F-MRI(functional magnetic resonance imaging)
5.PET (positronemitting tomography)
6.F-NIRS(functional near infrared spectroscopy)