Technology Used to Image the Brain - dummies

Technology Used to Image the Brain

By Frank Amthor

On the physiological front, electroencephalographic (EEG) recordings in the early 20th century revealed that the brain was constantly producing electrical oscillations that could be recorded from the surface of the skull. These oscillations changed with external stimulation and when different thought patterns occurred inside the brain. The EEG has a few deficiencies, though, which make it less than ideal for careful brain study.

  • The EEG averages activity over large areas so localizing the source of the recorded activity in the brain is hard. The averaging also tends to miss much of the complexity of brain activity.
  • The intrinsic ongoing activity in the brain tends to be larger than — and tends to swamp — the smaller, transient signal elicited by particular stimuli, such as showing someone a picture, which makes the EEG a poor method to study the effect of stimuli on brain activity.

This problem was solved to some extent by repeating the stimulus numerous times and recording each event so that, after hundreds of trials, the responses to the stimulus would be added together to increase their strength. Such recordings are generally called ERPs (Evoked Response Potentials). In the visual system, they are typically called VEPs (visual evoked potentials).

In the early to mid-20th century, microelectrodes began to be used in animal experiments to reveal how single neurons produced action potentials. In the visual system, for example, a researcher could follow the signal from the photoreceptor’s electrical response to light, through the retina, and on to the thalamus and visual cortex. Similarly, on the motor side, muscle action potentials, alpha motor neuron spikes, and motor cortex firing were recorded with microelectrodes. However, because microelectrode recording is invasive (you have to insert microelectrode needles near the neurons to get a recording), it has had little or no applicability in humans.

After the middle of the 20th century new imaging techniques began to be commonly used.

PET and SPECT

PET stands for Positron Emission Tomography. In this technique, short-term weakly radioactive oxygen or sugar molecules are introduced into patients (or volunteers) while they perform some cognitive task. The most active neurons take up the radioactive substance because of their higher metabolism, and the locus of radioactivity in the brain is ascertained by a complex scanner and sophisticated software that detects the emission of positrons (anti-electrons).

Typically a scan is done at “rest” — before the cognitive task is performed — and then compared with another scan done while the subject is doing the task. In this way, the difference between the two scans is assumed to be task associated. The PET technique has a spatial resolution of millimeters, but because it takes only one snapshot of brain activity, it can’t resolve brain activity changing in time.

SPECT (Single Photon Emission Computed Tomography) is similar to PET except the ingested radioactive emitter releases gamma rays that are detected by the scanner.

fMRI

The fMRI (functional Magnetic Resonance Imaging) is a derivative of MRI, which is used to image brain structure at high resolution. The MRI imaging technique uses a high magnetic field and radio frequency to detect transitions in the spin of protons (usually in water molecules) in the brain, and was formerly referred to as nuclear magnetic resonance (NMR).

Reconstructed images have good contrast between areas of the brain with high neural cell body density (gray matter) versus areas consisting mostly of fiber tracts (white matter), at millimeter resolution. In fMRI, blood flow or blood oxygenation/deoxygenation is detected dynamically with a time resolution of several seconds (future instruments may even be able to do this even faster). Blood flow changes and blood deoxygenation are metabolic measures believe to reflect real neural activity.

Although the spatial resolution of fMRI is about an order of magnitude worse than MRI in the same instrument, typically a structural MRI scan is done first and then the fMRI functional scan is superimposed on the structural scan to locate areas of differential activity between task and rest. Because it uses no radioactive isotopes, fMRI is considered safe and protocols can be repeated many times in a single session to improve signal to noise.

MEG

MEG (magnetoencephalography) is based on the physics of electrical and magnetic fields. Active neurons generate electric currents that flow within the neurons and in the surrounding extracellular space. The EEG detects these currents after they have become diffused across extracellular space, including the skull between the electrodes and the brain, leading to poor spatial resolution.

A fundamental property of electric currents, however, is that they produce magnetic fields, which aren’t shunted by the intervening volume electrical conductor. The neural magnetic fields are extremely small, though, so MEG technology requires heavily shielded recording rooms and exotic low strength magnetic field detectors called SQUIDs (super-conducting quantum interference devices). Current research and medical instruments use many of these SQUIDs over the entire brain to image a large region of the brain simultaneously. MEG has temporal resolution nearly as fast as EEGs, and spatial resolution on the order of fMRI.

DTI

DTI (diffusion tensor imaging) is a variant of magnetic resonance imaging, and is sometimes called diffusion MRI. It measures the movement (diffusion) of water in the brain. Because water moves most easily down the cytoplasm of long, cylindrical neural axons, DTI effectively measures axon tracts. It is an important clinical instrument because axon tracts may be destroyed by strokes or other brain disorders.

Optical imaging

Optical imaging is a promising future technology that involves either the changes in absorption or scattering of light by neurons due to their electrical activity, or the use of reporter (typically fluorescent) dyes that respond to changes in the concentration of certain ions (like calcium) entering the neuron during activity. Optical techniques have been used extensively in experiments in isolated tissue obtained from animals. Changes in scattering or absorption of infrared light have been used to a lesser extent in humans where optical access to tissue has existed, such as retina or cortex during surgery. There are also techniques to use infrared light obtain EEG-like recordings optically through the skull.

More recently, neuroscience has exploded with optogenetics, which uses genetic modification to produce neurons that change fluorescence when they’re active or can be activated or inhibited when illuminated. The ability to make transgenic animals with optogenetic reporters or stimulators expressed only in particular neurons in particular areas is revolutionizing brain study.