MRI Pulse Sequences: Spin Echo, Fast Spin Echo, Gradient Echo, and Echo Planar Imaging

Overview

The four major MRI pulse sequence families, spin echo (SE), fast spin echo (FSE/TSE), gradient echo (GRE), and echo planar imaging (EPI), represent distinct strategies for sampling k-space, with consequent differences in contrast, acquisition speed, SNR, and artefact profile. Mastery of their principles, trade-offs, and clinical applications is essential for the RANZCR Part 1 examination.

Spin Echo (SE) Sequences

Mechanism

Each TR interval uses two RF pulses:

1. A 90° excitation pulse tips longitudinal magnetisation into the transverse plane.
2. A 180° refocusing pulse at TE/2 rephases transverse dephasing, forming a spin echo at TE.

The 180° pulse compensates for static $B_0$ inhomogeneities; signal therefore decays according to true T2 (not the faster T2*). This makes SE inherently robust against susceptibility and field-uniformity artefacts.

Acquisition Time

$$\text{Scan time} = TR \times N_{PE} \times NSA$$

where $N_{PE}$ is the number of phase-encoding steps and $NSA$ (NEX) is the number of signal averages. A typical T2-weighted SE acquisition (TR 2500 ms, 256 phase steps, 1 average) takes ~10-11 minutes, practically prohibitive without multi-slice interleaving. Standard multi-slice SE acquisitions require 10-20 minutes.

Image Contrast

Advantages and Limitations

Advantages: True T2 contrast unaffected by field inhomogeneity; high SNR; reliable, reproducible tissue contrast; gold standard for contrast comparison.

Limitations: Long acquisition times; prone to motion artefact over prolonged scanning; unsuitable for breath-hold or dynamic acquisitions. Largely replaced in routine practice by FSE for T2-weighted imaging and GRE for T1-weighted body imaging.

Fast Spin Echo (FSE) / Turbo Spin Echo (TSE)

Mechanism

FSE (synonyms: TSE, RARE, echo-train SE) applies multiple 180° refocusing pulses within a single TR, each with a different phase-encoding gradient, filling multiple k-space lines per excitation. The number of echoes per TR is the echo train length (ETL) or turbo factor (TF).

$$\text{Scan time}{FSE} = \frac{TR \times N{PE} \times NSA}{ETL}$$

An ETL of 8 reduces scan time eightfold. Combined with half-Fourier (conjugate symmetry) techniques, acquisition time can be halved again. The effective TE is determined by the echo filling the centre of k-space, which dominates image contrast.

Image Contrast Compared with Conventional SE

Fat Brightness, Critical Pitfall

Fat appears paradoxically bright on FSE T2-weighted images relative to conventional SE. The rapid train of 180° pulses disrupts J-coupling between adjacent methylene protons in lipid molecules, preventing the normal T2 shortening of fat. Consequences:

• Vertebral metastases may be obscured in fatty bone marrow on FSE T2 images.
• Perilesional oedema may be masked adjacent to fat.
• Fat suppression (chemical saturation or STIR) must be added when evaluating marrow pathology, soft-tissue lesions, or perilesional oedema.

Image Blurring

Signal amplitude decays via T2 across the echo train. Later echoes, filling peripheral k-space, encoding fine spatial detail, have lower amplitude, producing blurring that worsens with longer ETL and shorter tissue T2. This is the fundamental image quality trade-off of FSE: longer ETL = faster scan + more blurring + higher SAR.

FSE Variants

STIR suppresses all tissues with short T1 (including fat) and accentuates high-water-content tissues; it more closely resembles a strongly T2-weighted image and is more reliable than spectral fat saturation near metal or at field inhomogeneities.

Gradient Echo (GRE) Sequences

Mechanism

GRE sequences omit the 180° refocusing pulse. Signal formation uses gradient reversal: a dephasing gradient followed by a rephasing (frequency-encoding) gradient reconstitutes the echo. Consequences:

• $B_0$ inhomogeneities are not compensated → signal decays by T2* (much shorter than T2).
• A partial flip angle ($\alpha < 90°$) allows shorter TR without full saturation of longitudinal magnetisation, enabling very fast acquisitions.

The Ernst angle defines the flip angle giving maximum SNR for a given TR and T1:

$$\cos \alpha_{Ernst} = e^{-TR/T1}$$

Low flip angles (10-30°) with short TR minimise T1 weighting; higher flip angles (60-90°) increase T1 weighting. Image contrast is determined by the combination of flip angle, TR, and TE.

Image Contrast

T2* contrast is unique to GRE and provides sensitivity to:

• Haemorrhage at all stages (deoxyhemoglobin, methemoglobin, haemosiderin, ferritin shorten T2*)
• Iron deposition (hepatic iron overload, haemochromatosis, transfusional iron)
• Calcification
• Air-tissue interfaces

Flowing blood appears bright in GRE (flow-related enhancement), basis of time-of-flight MR angiography. On conventional SE sequences, flowing blood produces a flow void (low signal); on GRE it is typically bright. This distinction is diagnostically important (e.g. differentiating phlebolith from flow void in vascular malformations).

GRE Variants

Balanced SSFP (bSSFP): Symmetrical gradients in all three directions accumulate both FID and stimulated echo signals, yielding T2/T1 contrast with high SNR and very fast acquisition. Particularly useful for cardiac cine and dynamic imaging. Uniquely prone to banding artefacts at $B_0$ inhomogeneities (constructive/destructive interference of coherent magnetisation).

Advantages and Limitations

Advantages: Very fast (TR 2-200 ms); enables breath-hold and dynamic acquisitions; bright blood without contrast; T2* sensitivity for haemorrhage, iron, calcification; T1-weighted GRE has replaced SE T1 for body imaging.

Limitations: T2* weighting (not true T2) → exaggerated susceptibility artefact near metal, bone interfaces, air; lower soft-tissue contrast than SE; banding artefacts (bSSFP); chemical shift artefact more prominent at longer TE.

Echo Planar Imaging (EPI)

Mechanism

EPI is the fastest clinically available MRI technique, acquiring a complete image in 20-100 ms after a single RF excitation (single-shot EPI). After excitation, an oscillating readout gradient rapidly alternates polarity, generating a rapid train of gradient echoes; simultaneous small phase-encoding gradient "blips" step through k-space one line at a time in a zigzag pattern until k-space is filled.

The effective TE corresponds to the time when gradient echo amplitude peaks at the centre of k-space (where contrast information resides).

All k-space must be sampled within a time $\lesssim T2^*$ (~50 ms), placing extreme demands on:

• Gradient coils (shielded, low eddy-current coils required)
• Gradient switching speed and linearity
• RF transmitter/receiver hardware
• Analogue-to-digital conversion rate

EPI Implementations

SE-EPI: 90° excitation → phase/frequency encoding gradients to peripheral k-space → 180° refocusing pulse → oscillating gradient readout train. The 180° pulse reduces susceptibility effects and provides T2 weighting at effective TE. Longer acquisition time than GRE-EPI; higher SAR; better image quality.

GRE-EPI: Single excitation pulse only (no 180° pulse) → oscillating gradient readout. Faster; T2*-weighted; more susceptible to distortion and signal dropout.

GRASE (Gradient and Spin Echo): Hybrid sequence combining initial spin echo with a series of gradient echoes per TR, repeated with further 180° RF pulses until k-space is filled. Achieves speed of GRE with RF compensation of SE. Trade-off: longer acquisition (>100 ms) and greater SAR from multiple 180° pulses.

Image Quality

Clinical Applications

Artefacts in EPI

Safety Concerns with EPI

• Peripheral nerve stimulation (PNS): Rapidly switching gradients (high $dB/dt$) can stimulate peripheral nerves; IEC limits on gradient switching rates apply.
• Acoustic noise: Loud gradient switching generates noise >100 dB; hearing protection is mandatory.
• SAR: SE-EPI has higher RF energy deposition than GRE-EPI due to the 180° refocusing pulse.
• Eddy currents: Rapid gradient switching induces eddy currents in conductive structures; shielded gradient coils are required to minimise these.

Comparative Summary

RF Energy and Safety Considerations

MRI uses non-ionising radiofrequency radiation with no ionising radiation dose. RF energy deposition is quantified by the Specific Absorption Rate (SAR) in $\text{W kg}^{-1}$. IEC and ARPANSA limits typically apply as 2-4 $\text{W kg}^{-1}$ whole-body averaged over 6 minutes. FSE/TSE and SE-EPI, with their multiple 180° pulses, generate the highest SAR. At 3T, SAR is approximately four times higher than at 1.5T for equivalent sequences, necessitating sequence modifications: reduced ETL, extended TR, or variable-flip-angle refocusing pulses (e.g. hyperecho techniques).

$$\text{Scan time}{SE} = TR \times N{PE} \times NSA \qquad \text{Scan time}{FSE} = \frac{TR \times N{PE} \times NSA}{ETL}$$

High-Yield Examination Points

• The 180° refocusing pulse in SE/FSE compensates for static $B_0$ inhomogeneity → true T2 contrast and low susceptibility artefact. GRE lacks this → T2* weighting → amplified susceptibility effects.
• FSE fat brightness (T2WI) is a clinically important pitfall due to J-coupling suppression by rapid 180° pulses, always add fat suppression when evaluating bone marrow, perilesional oedema, or soft-tissue pathology adjacent to fat.
• GRE is the sequence of choice for detecting haemorrhage (all stages), iron deposition, and calcification due to T2* sensitivity. Flowing blood is bright on GRE (flow-related enhancement) but produces a flow void on SE.
• EPI enables DWI and fMRI; its characteristic artefacts (geometric distortion, N/2 Nyquist ghost, signal dropout) arise directly from rapid single-shot k-space traversal without RF compensation for field inhomogeneities.
• Increasing ETL in FSE reduces scan time but increases blurring, SAR, and fat signal brightness on T2WI.
• Balanced SSFP (bSSFP/TrueFISP/FIESTA) generates T2/T1 contrast with high SNR; ideal for cardiac cine; uniquely prone to banding artefacts from $B_0$ inhomogeneity.
• GRASE combines gradient echoes with spin echo RF refocusing, offering a speed/image quality compromise between EPI and FSE.
• Susceptibility artefact on GRE is demonstrated dramatically by comparison with FSE: the same dental filling or haemorrhagic lesion appears as a large signal void on GRE-T2* but is substantially suppressed on FSE-T2 due to 180° pulse compensation.

Sources

• RANZCR position papers & standards
• Radiopaedia (open educational resource)