Skip to content

MRI

  • Many atomic nuclei (notably hydrogen, abundant in the body as water and fat) have an intrinsic property called spin.
  • This spin gives each nucleus a tiny magnetic dipole moment.
  • In the absence of an external magnetic field, these dipoles are randomly oriented.

When a strong external magnetic field (B₀) is applied, the magnetic moments of nuclei align either:

  • Parallel (low energy state) with B₀
  • Antiparallel (high energy state) to B₀

The nuclei don’t just sit aligned; they precess around the B₀ direction at a specific frequency For clinical MRI (1.5–3 Tesla), this falls in the radiofrequency (RF) range.

  • If an RF pulse matching the Larmor frequency is applied, nuclei absorb energy and “tip” away from alignment with B₀.
  • This creates a transverse magnetization (in the x-y plane) that rotates at the Larmor frequency.

Once the RF pulse stops, nuclei relax back toward equilibrium:

  • T1 relaxation (longitudinal recovery):
    • Magnetization along B₀ regrows.
    • Time constant of T1 depends on tissue properties.
  • T2 relaxation (transverse decay):
    • Coherence of transverse magnetization is lost due to spin–spin interactions.
    • Time constant of T2 also tissue dependent and generally shorter than T1.

These differences in relaxation times allow MRI to distinguish tissues (e.g. gray vs white matter, healthy vs diseased tissue).

  • The changing transverse magnetization induces a current in receiver coils via Faraday’s law of induction.
  • This produces the measurable MR signal, which decays over time.

To form images, spatial information must be encoded:

  • Magnetic field gradients are applied along x, y, and z axes.
  • These gradients make the Larmor frequency vary with position, enabling localization of signal sources via Fourier transforms.

By choosing timing parameters (e.g., repetition time TR, echo time TE) and pulse sequences, MRI can emphasize:

  • T1 differences (good for anatomy)
  • T2 differences (good for detecting pathology like edema)
  • Proton density (water vs fat content)
  • More advanced contrasts (diffusion, perfusion, functional MRI).

There are a plethora of MRI sequences being used clinically to assist with diagnosis.

SequenceDescriptionTR (msec)TE (msec)
T1-WeightedShort TR and TE50014
T2-WeightedLong TR and TE400090
T2-FLAIRVery Long TR and TE9000114
T2*T2* decay makes detection of magnetic susceptibility effects such as calcium and blood easier
Steady State Sequences- Steady-state free procession (SSFP)
- Fast imaging employing steady-state acquisition (FIESTA)/Constructive interference in steady-state (CISS)
DixonCan generate four sequences. Can be used for fat suppression.
Turbo Spin Echo
Gradient Echo
Gradient EchoSpin Echo
FasterSlower
Flip angle <90 degrees (10-80)Flip angle 90 at or close to degrees

In original method, two sets of spin echo images were acquired with slightly different echo times and four sets of images are generated

  • water only
  • fat only
  • in-phase (IP)
  • out-of-phase (OOP)
ProsCons
- Generally better than CHESS/Fat-sat sequences- Fat-water swap
- Limitations around highly inhomogeneous areas such as neck and around metal

See descriptions of MRI artifacts along with cases at Radioepaedia.

  1. Whats the difference between T2* decay vs. T2 decay? See MRI relaxaton curves.
  1. https://mriquestions.com/index.html
  2. https://mrimaster.com/
  3. https://radiopaedia.org/home