Home / Exam guides / MRI physics made simple

MRI physics made simple

The ten ideas you actually need: how the signal is made, T1 vs T2, blood ageing, sequences, artifacts and safety.

By Dr AbdullahUpdated 1 November 202611 min read

MRI physics scares most residents because it is usually taught as equations. You do not need the equations to read scans or to pass FRCR physics: you need about ten ideas, explained simply. Here they are.

How an MR image is made, in five steps

  1. Protons line up. Hydrogen nuclei in water and fat behave like tiny magnets and align with the scanner's main field (B0), creating net magnetisation along the field.
  2. They precess at the Larmor frequency, which is proportional to field strength: about 42.58 MHz per tesla, so about 64 MHz at 1.5 T and 128 MHz at 3 T.
  3. An RF pulse tips them over into the transverse plane, where their rotating magnetisation induces a signal in the receiver coil.
  4. They relax. Longitudinal magnetisation recovers (T1 recovery) and transverse magnetisation decays as spins dephase (T2 decay).
  5. Gradients encode position (slice selection, phase encoding and frequency encoding), filling k-space, which a Fourier transform turns into the image.

T1, T2 and PD weighting

WeightingTRTEFluid looksGood for
T1ShortShortDarkAnatomy, fat, contrast enhancement, subacute blood
T2LongLongBrightOedema, fluid, most pathology
Proton densityLongShortIntermediate to brightJoints, menisci and cartilage

Memory aid: TR controls T1 weighting, TE controls T2 weighting. Short TR exaggerates T1 differences; long TE exaggerates T2 differences.

TissueT1T2
Water/CSFDarkBright
FatBrightBright on fast spin echo
Cortical bone, air, calcificationDarkDark
Gadolinium-enhancing tissueBrightLittle effect at clinical doses
Melanin, protein-rich fluid, methaemoglobinBrightVariable

Blood on MRI: the age table

StageTimeMain productT1T2
HyperacuteUnder 24 hoursOxyhaemoglobinIsoBright
Acute1–3 daysDeoxyhaemoglobinIsoDark
Early subacute3–7 daysIntracellular methaemoglobinBrightDark
Late subacute1–4 weeksExtracellular methaemoglobinBrightBright
ChronicMonthsHaemosiderinDarkDark (rim)

The classic mnemonic reads the T1 then T2 pattern: "It Be, Iddy Biddy, Baby, Doo Doo" (I = iso, B = bright, D = dark).

The sequences you will meet

MRI artifacts: cause and fix

ArtifactCauseFix
Motion/ghostingMovement, breathing, pulsation; appears along the phase-encoding directionBreath-hold, gating, swap phase and frequency directions, saturation bands
Aliasing (wrap-around)Anatomy outside the field of view in the phase directionLarger field of view, oversampling, swap directions
Chemical shiftFat and water precess at slightly different frequencies; dark and bright bands at fat–water edges along the frequency directionIncrease receiver bandwidth, fat suppression
SusceptibilityMetal, air or blood products distort the local fieldSpin echo instead of gradient echo, higher bandwidth, metal artifact reduction sequences
Gibbs (truncation)Too few phase-encoding steps at sharp boundaries; fine parallel lines (can mimic a syrinx)Increase matrix size
ZipperRadiofrequency leak, for example a door left openClose the door, check the RF shielding
Magic angleTendons at about 55° to B0 look bright on short TE sequencesCheck on a long TE (T2) sequence

MRI safety in one minute

1.5 T versus 3 T

3 T roughly doubles the signal-to-noise ratio, which buys thinner slices or faster scans, and it helps neuro, MSK and prostate imaging. The trade-offs are more susceptibility and chemical shift artifact, higher SAR, more standing-wave (dielectric) artifact in the abdomen and more implants with restrictions. Many body and cardiac studies are still done well at 1.5 T.

Ready to test yourself? Try our 20 free FRCR physics true/false questions, which include MRI.

Frequently asked questions

What is the Larmor frequency at 1.5 T and 3 T?

Hydrogen precesses at about 42.58 MHz per tesla, so about 64 MHz at 1.5 T and about 128 MHz at 3 T.

How do I tell T1 from T2 images quickly?

Look at fluid such as CSF, urine or joint fluid: dark on T1, bright on T2. Fat is bright on both T1 and fast spin echo T2, so use fluid, not fat.

What is the difference between STIR and FLAIR?

Both are inversion recovery sequences. STIR uses a short inversion time to null fat; FLAIR uses a long inversion time to null CSF.

What is T2 shine-through?

A lesion that is bright on DWI because it is very bright on T2, not because diffusion is restricted. The ADC map is bright, not dark, which tells you diffusion is not restricted.

Written by the Radiology Boards team from standard MRI physics teaching. Safety rules summarised; always follow your department's MR safety policy.