GloNeuro

Neuroimaging

Neuroimaging shows the structure and activity of the living brain without surgery, in health and in disease.

SubjectBrain structure, activity and chemistry
ScaleMillimetres to the whole brain
Applied toPatients and volunteers, in hospitals and research centres

Methods by scale

Microscopy
about 1 nm to 1 cm
Neuroimaging
about 0.5 mm to the whole brain
Brain networks
single neurons to the whole brain
Electrophysiology
about 10 nm to the whole brain
1 nm10 nm100 nm1 µm10 µm100 µm1 mm1 cm10 cm1 m
  1. Ion channelabout 10 nm
  2. Synaptic vesicleabout 40 nm
  3. Neuron cell bodyabout 10–30 µm
  4. Thickness of cortexabout 2–4 mm
  5. Human brainabout 15 cm long
Spatial scale, logarithmic: each step along the axis is ten times larger than the last. Ranges and sizes are approximate.
From neural activity to the fMRI signal. Active neurons trigger a local increase in blood flow that exceeds their extra oxygen use, so the proportion of oxygenated blood rises. fMRI detects this blood-oxygen-level-dependent (BOLD) change, which peaks about 5 seconds after the activity and is often followed by a small undershoot. The curve is the standard model of the haemodynamic response.

Structural imaging produces detailed pictures of the brain’s anatomy: grey and white matter, the fluid-filled ventricles, and abnormalities such as tumours, bleeding or tissue damaged by stroke.

Functional imaging measures brain activity indirectly. Functional MRI tracks changes in blood oxygenation that follow neural activity, while positron emission tomography (PET) uses small amounts of radioactive tracers to measure metabolism, blood flow or specific molecules, such as the amyloid deposits found in Alzheimer’s disease.

Because these methods are non-invasive or minimally invasive, they make it possible to study the human brain across the lifespan, to diagnose disease, and to plan and monitor treatment.

Key techniques

Magnetic resonance imaging (MRI)

Detailed pictures of brain anatomy without ionising radiation.

A strong magnetic field aligns hydrogen nuclei, mostly in water molecules. Radio-frequency pulses tip them out of alignment, and as they relax they emit signals that differ between tissues. By varying the timing of the pulses, scanners create images in which grey matter, white matter and fluid look different, or that highlight swelling, scarring or bleeding.

Applications include diagnosing tumours, multiple sclerosis and stroke, measuring brain volume and cortical thickness and planning neurosurgery.

Most clinical scanners operate at 1.5 or 3 tesla, tens of thousands of times stronger than the Earth’s magnetic field.

Functional MRI (fMRI)

Mapping brain activity through changes in blood oxygenation.

Active neurons need more energy, and local blood flow increases more than oxygen use. This changes the balance of oxygenated and deoxygenated haemoglobin, which have different magnetic properties. fMRI detects this blood-oxygen-level-dependent (BOLD) signal, which peaks a few seconds after activity. Comparing scans during different tasks, or at rest, shows which regions are engaged.

Applications include mapping language and motor areas before surgery, studying cognition, emotion and decision-making and measuring functional connectivity at rest.

fMRI measures an indirect, delayed signal: it shows where blood flow changes, not neural firing itself.

Diffusion MRI and tractography

Tracing the brain’s white-matter pathways.

Water molecules diffuse more easily along nerve fibres than across them. Diffusion MRI measures this directional movement throughout the brain. Tractography algorithms follow the preferred direction from point to point to reconstruct likely fibre pathways, such as the corticospinal tract or the arcuate fasciculus.

Applications include assessing white-matter injury after trauma or stroke, locating pathways to avoid during surgery and studying brain development and ageing.

Tractography infers pathways from the movement of water; it does not trace individual axons, and crossing fibres can cause errors.

Computed tomography (CT)

Fast X-ray imaging, essential in emergencies.

An X-ray source and detectors rotate around the head, measuring how much radiation passes through from many angles. A computer reconstructs cross-sectional images in which bone appears bright and fluid dark. Fresh blood is denser than brain tissue, so bleeding stands out clearly.

Applications include detecting haemorrhage in suspected stroke or head injury, showing skull fractures and guiding urgent treatment decisions.

A CT scan takes seconds and is widely available, but it uses ionising radiation and shows soft tissue less clearly than MRI.

Positron emission tomography (PET)

Measuring metabolism and specific molecules with radioactive tracers.

A small amount of a tracer labelled with a positron-emitting isotope is injected. When a positron meets an electron, both are annihilated, releasing two gamma rays that travel in opposite directions. Detectors around the head record these pairs, and the tracer’s distribution is reconstructed. Fluorodeoxyglucose shows glucose use; other tracers bind amyloid, tau or neurotransmitter receptors.

Applications include detecting amyloid and tau in suspected Alzheimer’s disease, locating the source of seizures and studying dopamine systems in Parkinson’s disease and addiction.

PET is usually combined with CT or MRI so that the tracer signal can be placed on the anatomy.

Functional near-infrared spectroscopy (fNIRS)

Light-based measurement of activity in the outer cortex.

Near-infrared light passes through the scalp and skull and scatters within the brain. Oxygenated and deoxygenated haemoglobin absorb these wavelengths differently, so detectors placed a few centimetres from the light sources measure changes in blood oxygenation in the cortex beneath.

Applications include studying brain development in infants, recording activity during natural movement and conversation and monitoring at the bedside.

fNIRS is quiet, portable and tolerant of movement, but it reaches only the outer layer of the cortex.

Comparing methods

MethodWhat it measuresSpatial resolutionTemporal resolution
Structural MRITissue anatomyAbout 1 mmMinutes per scan
fMRIBlood oxygenation linked to activityAbout 1–3 mmSeconds (the blood response lags activity)
PETDistribution of a tracerA few millimetresMinutes
CTX-ray density of tissueUnder 1 mmSeconds

Values are approximate and vary with the scanner and protocol.

The same patch of folded cortex at the typical voxel size of each scan, across a 24 mm field. Finer voxels resolve the folds of grey matter; at the voxel sizes of fMRI and PET, grey matter, white matter and fluid blur together. Contrast is kept the same in every panel so that only voxel size differs; in practice CT shows grey and white matter with much less contrast than MRI, and real resolution depends on the scanner and protocol.

History of the field

  • Discovery
  • Technique
  • Application

  1. 1927Technique

    Cerebral angiography

    Egas Moniz makes the brain’s blood vessels visible on X-ray images.

  2. 1971Technique

    The first CT scan of the brain

    Godfrey Hounsfield’s scanner produces cross-sectional X-ray images of a patient’s head.

  3. 1970sTechnique

    Positron emission tomography

    PET allows brain metabolism and blood flow to be measured in living people.

  4. 1980sTechnique

    MRI enters the clinic

    Magnetic resonance imaging provides detailed anatomy without ionising radiation.

  5. 1990Discovery

    The BOLD signal

    Seiji Ogawa and colleagues show that blood oxygenation changes MRI contrast, the basis of functional MRI.

  6. 1992Technique

    Functional MRI in humans

    The first studies map human brain activity during simple tasks.

  7. 1994Technique

    Diffusion tensor imaging

    Measuring the direction of water diffusion opens the way to tracing white-matter pathways.

  8. 1995Discovery

    Resting-state connectivity

    Spontaneous fMRI signals are found to be correlated between distant motor regions, revealing networks at rest.

  9. 2004Application

    Imaging amyloid

    PET with Pittsburgh compound B makes the amyloid plaques of Alzheimer’s disease visible in living patients.

  10. 2017Application

    7 tesla in the clinic

    The first 7-tesla MRI scanner is cleared for clinical use in the United States, offering finer anatomical detail.

  11. 2020Application

    Imaging tau

    A PET tracer for tau, the second hallmark protein of Alzheimer’s disease, is approved for clinical use in the United States.

  12. 2024Technique

    11.7 tesla

    The Iseult scanner in France produces its first images of the human brain at a field strength of 11.7 tesla.

Strengths and limitations

StrengthsLimitations
Non-invasive study of the living human brainCannot resolve individual neurons
Covers the whole brain in a single sessionfMRI measures blood flow, an indirect and delayed sign of activity
Essential for diagnosis, surgical planning and monitoringScanners are expensive, and PET involves radiation exposure

Last updated