Turbo Spin Echo (TSE/FSE) Sequence

MRINinja Sequence Library | Child Page Parent page: 9003 — MRI Sequences: Overview and Classification Document Type: Sequence-Specific Technical Reference Technical Level: B — Protocol Optimisation Specialist Version: 1.0 — May 2026

1. Introduction: Historical Evolution and Clinical Purpose

The Turbo Spin Echo (TSE) sequence — universally known as Fast Spin Echo (FSE) in GE and Canon/Hitachi terminology — represents the single most important technical advance in clinical MRI acquisition strategy since the introduction of the original Spin Echo (SE) by Hahn (1950) [E1] and its clinical adaptation in the 1970s and 1980s.

The fundamental problem with conventional SE was temporal: a complete T2-weighted brain examination required 10–20 minutes of acquisition time per plane, making the technique clinically impractical at high resolution and precluding widespread deployment. The solution arrived in 1986, when Jürgen Hennig and colleagues at Freiburg published the RARE (Rapid Acquisition with Relaxation Enhancement) technique [E2], demonstrating that multiple 180° refocusing pulses could be applied sequentially after a single 90° excitation, with each refocused echo filling a different k-space line. The net effect was that scan time was reduced by a factor equal to the number of echoes acquired per excitation — the Echo Train Length (ETL), also called the Turbo Factor. An ETL of 8 reduces acquisition time by 8-fold; an ETL of 16 by 16-fold.

The clinical implications were transformative. TSE made high-resolution T2-weighted imaging routine, replaced conventional SE in virtually every anatomical application, and established itself as the backbone of modern clinical MRI. Today, the majority of all T2-weighted acquisitions worldwide — brain FLAIR, spine T2, pelvic T2, musculoskeletal cartilage — are TSE or TSE-derived sequences.

TSE was not designed as a replacement for all SE applications indiscriminately. The multi-echo acquisition introduces specific contrast modifications — most notably the brightening of fatty tissue due to stimulated echo effects and J-coupling modulation — that distinguish TSE contrast from true SE contrast in important ways. These differences are clinically relevant in musculoskeletal imaging, hepatic imaging, and any protocol where accurate fat signal behaviour is diagnostic.

The current clinical role of TSE encompasses five broad functions: (1) primary T2 contrast generation in virtually all anatomical regions; (2) T1-weighted acquisition at short ETL with TR optimisation; (3) foundation for Proton Density (PD)-weighted imaging in musculoskeletal protocols; (4) basis for all FLAIR sequences via inversion recovery preparation; and (5) platform for 3D isotropic acquisitions via Variable Refocusing Flip Angle TSE (SPACE/CUBE/VISTA family). These applications are developed in Section 6 and in dedicated child pages.

2. Physical Foundations

2.1 Pulse Sequence Logic

TSE begins identically to conventional SE: a 90° RF excitation pulse rotates longitudinal magnetisation (Mz) into the transverse plane, generating transverse magnetisation (Mxy). At this point, conventional SE applies a single 180° refocusing pulse and acquires one echo per TR period.

TSE instead applies a train of N successive 180° refocusing pulses within the same TR period. Each 180° pulse regenerates a spin echo by reversing phase dispersion caused by static field inhomogeneities. Each resulting echo is acquired with a different phase-encoding gradient, filling a different line in k-space. The number of 180° pulses — and therefore the number of k-space lines acquired per TR — defines the ETL (Echo Train Length) or Turbo Factor.

Because k-space lines are acquired at different time points relative to the initial 90° excitation, each echo carries a different degree of T2 weighting. The echo assigned to the center of k-space (which dominates image contrast and spatial uniformity) determines the effective TE (eff. TE). Echoes assigned to the periphery of k-space (acquired at other time points in the train) contribute mainly to fine spatial detail without substantially altering the dominant tissue contrast.

The fundamental consequence is that TSE is not a single-TE acquisition: it is a multi-TE acquisition in which contrast is governed by the eff. TE, but in which T2-related amplitude modulation of k-space lines introduces specific image quality effects — particularly edge blurring and fat signal alteration — that are absent from true SE.

2.2 Signal Formation and Contrast Generation

The signal in a TSE sequence follows the same fundamental CPMG (Carr-Purcell-Meiboom-Gill) [E3, E4] multi-echo logic:

Signal at echo n:

Sn∝ρ⋅(1−e−TR/T1)⋅e−TEn/T2S_n \propto \rho \cdot \left(1 - e^{-TR/T1}\right) \cdot e^{-TE_n/T2}Sn​∝ρ⋅(1−e−TR/T1)⋅e−TEn​/T2

where ρ is the proton density, T1 governs longitudinal recovery between TR periods, and T2 governs transverse decay along the echo train. The eff. TE is the TE of the echo placed at k-space center.

Three contrast regimes are achievable:

• T1-weighted TSE: Short TR (300–700 ms at 1.5T; 400–800 ms at 3T), short eff. TE (10–25 ms), short ETL (2–4). Suitable for T1 contrast, though true SE remains preferred for highest T1 accuracy in musculoskeletal applications.
• Proton Density (PD)-weighted TSE: Long TR (>2000 ms), short eff. TE (15–30 ms), moderate ETL (4–8). Dominant application: articular cartilage, menisci, ligaments.
• T2-weighted TSE: Long TR (2500–5000 ms), long eff. TE (80–120 ms), moderate-to-long ETL (8–24). Dominant application: brain, spine, abdomen, pelvis.

The fat signal in TSE — a key distinction from SE:

In conventional SE, fat behaves as a short-T2 species relative to long-T2 tissues. In TSE, the rapid succession of 180° pulses suppresses J-coupling-driven T2 shortening of methylene protons in long-chain fatty acids. This restores fat signal along the echo train, causing fat to appear brighter on TSE T2-weighted images than it would on equivalent SE acquisitions [D1]. This effect is clinically important in musculoskeletal imaging (where fat within bone marrow and soft tissue must be accurately assessed) and is the main reason why TSE T2 without fat suppression cannot be directly equated to SE T2 in tissues with significant fat content.

2.3 Pulse Diagram

[Pulse Diagram — TSE ETL=4]

(SVG graphic — separate asset: TSE_pulse_diagram_ETL4.svg)

Five-channel display: RF pulses | Slice-selection gradient (Gz) | Phase-encoding gradient (Gy) | Frequency-encoding gradient (Gx) | Echo signal.

RF channel: 90° excitation (teal) followed by four 180° refocusing pulses (gold), each separated by the Echo Spacing (ES).

Echo channel: four echoes of decreasing amplitude following T2 decay envelope; Echo 2 marked as k₀ (k-space center) = effective TE.

Annotated: ES (Echo Spacing), eff. TE, TR. ETL=4 label and scan time formula at base.

Turbo Spin Echo / Fast Spin Echo pulse sequence timing diagram

3. Key Parameters and Their Clinical Meaning

3.1 Parameter Table

3.2 Temporal Magnetisation Diagrams

[Four magnetisation diagrams — separate assets]

Each diagram displays time on the horizontal axis across a minimum of 4 TR cycles.

Five tissue curves per diagram: Fat (orange), White Matter (blue), Grey Matter (teal), Muscle (red), CSF/Free Water (purple).

Upper sub-panel: Mz (longitudinal magnetisation recovery after each 90° pulse).

Lower sub-panel: Mxy (transverse magnetisation decay; final signal amplitude at echo sampling time).

White background, MRINinja palette.

Diagram 1 — T1-weighted, 1.5T (TSE_Mz_T1_1p5T.svg)

TR=500 ms, eff.TE=15 ms, ETL=3.

At TR=500 ms, longitudinal recovery is incomplete for all tissues at 1.5T. Fat recovers fastest (T1~260 ms) → brightest. White matter (T1~800 ms) recovers more than grey matter (T1~1100 ms) → moderate T1 contrast. CSF (T1~3500 ms) shows minimal recovery → lowest signal. The short eff.TE minimises T2 decay differences. The short ETL limits J-coupling fat brightening.

Caption T1-weighted:

Following the 90° excitation, Mz is nulled and begins recovering along an exponential curve governed by tissue T1. At TR=500 ms, each tissue has recovered a different fraction of its equilibrium Mz. Fat, with the shortest T1, has recovered ~86% of M0. White matter (~46%) and grey matter (~37%) show intermediate recovery with detectable difference — this is the source of T1 contrast. Muscle (~60%) and CSF (<14%) complete the dynamic range. At the moment of the echo (eff.TE=15 ms), T2 decay is minimal for all tissues — Mxy differences are negligible at 15 ms. The tissue signal hierarchy on the final image therefore directly reflects T1 recovery: Fat > Muscle > White Matter > Grey Matter >> CSF.

Diagram 2 — T2-weighted, 1.5T (TSE_Mz_T2_1p5T.svg)

TR=3000 ms, eff.TE=90 ms, ETL=16.

At TR=3000 ms and 1.5T, longitudinal recovery is nearly complete for all tissues except CSF. The contrast is dominated by Mxy decay at eff.TE=90 ms: CSF (T2~1400 ms) retains high signal; grey matter (T2~95 ms) retains moderate signal; white matter (T2~80 ms) slightly less; muscle (T2~35 ms) low signal; fat — despite short true T2 — appears bright due to J-coupling suppression along the echo train (stimulated echo effect, TSE-specific).

Caption T2-weighted:

At the long TR used for T2-weighted TSE, longitudinal recovery is essentially complete before the next excitation for all tissues except CSF. Contrast is therefore almost entirely determined by T2 decay during the echo train. At eff.TE=90 ms, CSF retains the highest signal (long T2 → slow decay). Grey matter retains slightly more signal than white matter due to marginally longer T2. Muscle shows rapid T2 decay → low signal. Fat behaves anomalously in TSE: the CPMG refocusing train suppresses J-coupling-mediated T2 shortening of methylene protons, maintaining fat signal at levels higher than expected from its intrinsic T2 — this is the TSE stimulated echo fat effect and must be considered when evaluating fat-containing lesions on non-fat-saturated T2 TSE images.

Diagram 3 — T1-weighted, 3T (TSE_Mz_T1_3T.svg)

TR=650 ms, eff.TE=15 ms, ETL=3.

At 3T, tissue T1 values are longer (fat~320 ms; WM~1000 ms; GM~1400 ms; CSF~4500 ms). At TR=650 ms, recovery curves are shifted rightward compared to 1.5T — absolute recovery fractions are lower. To maintain comparable T1 contrast, TR must increase relative to 1.5T (650–800 ms recommended). SAR implications from the ETL must be monitored.

Caption T1-weighted at 3T:

Tissue T1 relaxation times increase substantially at 3T compared to 1.5T. Fat elongates from ~260 ms to ~320 ms; white matter from ~800 ms to ~1000 ms; grey matter from ~1100 ms to ~1400 ms. At TR=650 ms, the differential recovery between tissues remains detectable but is slightly compressed relative to 1.5T at equivalent TR. To preserve T1 contrast, TR must be extended by approximately 100–200 ms compared with 1.5T protocols. The short eff.TE maintains minimal T2 contribution to contrast. SAR considerations at 3T with even a short ETL may require interleaved acquisition or slightly extended TR.

Diagram 4 — T2-weighted, 3T (TSE_Mz_T2_3T.svg)

TR=3500 ms, eff.TE=90 ms, ETL=16.

T2 values are largely field-strength-independent. The pattern is therefore nearly identical to the 1.5T T2 diagram. TR is extended to 3500 ms to accommodate longer T1 recovery. SAR at 3T with ETL=16 is substantially higher → variable flip angle refocusing (VFA-TSE) should be considered.

Caption T2-weighted at 3T:

Unlike T1, T2 relaxation times show minimal field-strength dependence. The signal hierarchy and decay pattern at eff.TE=90 ms at 3T is therefore essentially equivalent to 1.5T: CSF > Grey Matter > White Matter >> Muscle, with fat appearing anomalously bright due to the TSE stimulated echo effect. The TR extension to 3500 ms reflects the longer T1 at 3T, ensuring complete longitudinal recovery before the next excitation. The primary practical challenge at 3T with long ETL T2-weighted TSE is SAR management: the power deposited by 16 successive 180° pulses per TR at 3T can exceed SAR limits, requiring either VFA refocusing pulse strategies, extended TR, reduced ETL with extended echo spacing, or sequence-level SAR monitoring.

4. Tissue Contrast Profiles

5. Vendor Implementations

VendorTSE/FSE NameVFA-3D NameSingle-Shot NameKey Implementation NotesSiemensTSESPACEHASTEVFA SPACE uses hyperecho + variable flip angle train. SAR monitor integrated.GEFSECUBESSFSECUBE uses variable flip angle and optimised echo spacing. SSFSE with half-Fourier.PhilipsTSEVISTASS-TSEVISTA uses sinusoidal variable flip angle. mDixon integration available.CanonFSEPHASEFASEFASE (Fast Advanced SE) is single-shot; dedicated PHASE for 3D.HitachiFSEisoFSESS-FSELimited 3T product line; isoFSE available on higher-tier systems.

Key implementation differences:

SAR management at 3T: All vendors implement real-time SAR monitoring for long-ETL TSE. Siemens uses automatic TR extension or ETL reduction when SAR limits approach. GE and Philips implement similar safety algorithms. The specific SAR ceiling and monitoring strategy differ and are not user-visible without advanced console access.

Variable flip angle (VFA) strategies: The SPACE/CUBE/VISTA family uses different mathematically optimised flip angle trains. These are vendor-proprietary and are not interchangeable. Contrast behaviour at long ETL is similar but not identical across platforms.

Echo spacing: Minimum achievable echo spacing differs across vendors and is bandwidth- and ETL-dependent. Shorter ES reduces T2 blurring and fat J-coupling effects but increases gradient stress.

Parallel imaging integration: All vendors integrate TSE with their proprietary parallel imaging reconstruction (GRAPPA, ARC, SENSE). Reconstruction algorithms differ; SNR penalty and aliasing patterns are vendor-dependent.

AI reconstruction (Deep Learning Reconstruction, DLR): Available on recent platforms from all major vendors. Applied post-acquisition; does not alter the fundamental TSE k-space acquisition. SNR gain equivalent to 2–4× NEX increase has been reported for TSE sequences, enabling resolution increase or scan time reduction [D4].

6. Clinical Applications Overview

Clinical ApplicationAnatomical RegionPreferred RoleAlternative SequencesT2-weighted brain imagingBrain (all indications)Primary T2 sequenceSE (reference standard, now largely historical in routine)FLAIR (T2 + IR preparation)BrainWhite matter lesions, subarachnoid pathologyDIR (superior cortical lesion detection in MS)T2-weighted spine imagingCervical, thoracic, lumbar spinePrimary T2 sequence; disc, cord, CSFSTIR (add-on for marrow oedema)Pelvic organ imagingUterus, prostate, rectum, bladderPrimary soft tissue sequenceDWI (functional complement)Musculoskeletal — PD/T2Joints (knee, shoulder, hip, ankle)PD-weighted cartilage/ligament; T2 bone marrowTrue SE (gold standard for accurate T2 in MS cartilage)Liver and abdominal organsAbdomenT2 (breath-hold or free-breathing with navigator)ss-TSE HASTE (motion-robust alternative)Breast imagingBreastT2 characterisation of cysts, fibroadenomaDynamic T1 DCE (functional)Neck and salivary glandsHead and neckT2 soft tissue characterisationSTIR (nodal disease)Inner ear / IACPetrous pyramid3D TSE (CISS or SPACE for endolymph)CISS/FIESTA (CSF-fluid contrast)Fetal MRIFetal brain and bodyss-TSE (motion robustness)DWI (complementary)Post-contrast T1All regionsShort ETL T1-TSE (when SE not available)3D T1 GRE (VIBE/LAVA — preferred for volumetric coverage)MRCPBiliary tract, pancreatic ductHeavily T2-weighted TSE (eff.TE 600–1000 ms)3D MRCP-TSE with navigator

Strengths of TSE compared to conventional SE:

Acquisition time reduction by ETL factor; equivalent or superior CNR for most T2-weighted applications; full compatibility with parallel imaging and partial Fourier; readily combined with fat suppression techniques; basis for 3D isotropic acquisitions.

Limitations of TSE compared to SE:

Fat brightening on T2 (requires suppression for fat characterisation); T2 blurring at high ETL; SAR increase at 3T; slightly reduced spatial resolution at equivalent scan time unless parameters carefully optimised; mild susceptibility artefact reduction relative to GRE (may reduce T2* sensitivity).

7. Artefacts

8. Advanced Technical Parameters

8.1 ETL / Turbo Factor and Echo Train Design

The ETL is the single most consequential protocol variable in TSE after TR and eff.TE. Its effects are multiplicative: increasing ETL simultaneously reduces scan time, increases SAR, increases T2 blurring, increases fat J-coupling effect, and increases the number of k-space lines with progressively different T2 weightings per TR. Protocol optimisation therefore requires explicit ETL decisions rather than acceptance of scanner defaults.

For T1-weighted TSE, ETL should be kept at 2–4 in most applications. Beyond ETL=4, the fat J-coupling effect begins to alter T1 contrast unpredictably, and T1 weighting is diluted by T2 decay along the train. In post-contrast T1 applications, this matters because enhancing tissue contrast depends on precise T1 shortening — ETL-induced T2 weighting modifies the enhancement-to-background ratio.

For T2-weighted TSE, ETL of 8–24 is standard. The optimal ETL depends on TR, eff.TE, tissue composition, and field strength. At 3T, long ETL (>16) requires explicit SAR management. The minimum eff.TE achievable at a given ETL is determined by: eff.TE = (echo position within train) × ES. Shorter ES requires higher bandwidth, which reduces SNR per unit time — a fundamental trade-off.

8.2 Bandwidth

Receiver bandwidth (BW) per pixel determines: (1) chemical shift artefact magnitude (↑BW → ↓chemical shift), (2) minimum echo spacing (↑BW → shorter readout duration → shorter ES), (3) SNR (↑BW → ↓SNR by factor √(2·BW·matrix/BW_ref)). At 3T, a higher minimum BW is typically mandated to prevent chemical shift artefact from exceeding one pixel at fat-water interfaces. Standard operating range: 130–250 Hz/px at 1.5T; 200–400 Hz/px at 3T.

8.3 Parallel Imaging

All clinical TSE protocols should incorporate parallel imaging. GRAPPA (Siemens), ARC (GE), SENSE (Philips) are mathematically equivalent in principle but differ in implementation. Standard acceleration factor R=2 reduces scan time by ~50% with an SNR penalty of ~1/√2 (~29%). R=3 is feasible at 3T; R=4 requires careful coil coverage assessment and is mainly appropriate for 3D acquisitions with large coil arrays. Partial Fourier (6/8 phase fraction) provides additional ~25% scan time reduction; combining with R=2 can reduce total acquisition to ~35% of unaccelerated time.

8.4 3D TSE (SPACE/CUBE/VISTA)

Three-dimensional TSE acquisitions acquire a volumetric slab with isotropic or near-isotropic resolution, enabling retrospective multiplanar reformation. This eliminates the cross-talk limitation of 2D multislice TSE, allows thinner effective slice thickness, and improves partial volume suppression. The technical prerequisite is very long ETL (100–300+ echoes per TR), achieved by variable flip angle (VFA) refocusing trains that modulate the flip angle of successive 180° pulses. This dramatically reduces SAR while maintaining coherent signal through the echo train.

Key 3D TSE applications: brain morphology (1 mm³ T2 SPACE), spinal cord (0.8 mm³ isotropic), inner ear (CISS-competitive 0.4–0.6 mm³), MRCP (navigator-gated), liver lesion characterisation.

8.5 k-Space Ordering

Linear k-space ordering fills k-space sequentially from -kmax to +kmax. The eff.TE corresponds to the echo placed at k=0. This is the most common ordering for standard T2-weighted TSE.

Centric k-space ordering places the first echo (shortest TE) at k=0, then fills outward. This is preferred for T1-weighted TSE and contrast-enhanced TSE where maximal T1 contrast at minimal TE is desired, as it minimises T2 weighting at the contrast-determining center of k-space.

8.6 SAR Considerations at 3T

Specific Absorption Rate (SAR) scales with the square of the B1 field strength: SAR ∝ B1². At 3T, each 180° refocusing pulse deposits approximately 4× the power of the equivalent 1.5T pulse. A TSE sequence with ETL=16 and 30 slices at TR=3500 ms at 3T can approach or exceed IEC limits (2 W/kg head, 4 W/kg body) without mitigation. Standard mitigation strategies include: (1) VFA refocusing train (reduces average flip angle from 180° to 120–150°); (2) TR extension (reduces duty cycle); (3) ETL reduction with ES reduction; (4) reduced number of slices per slab. All major vendors implement real-time SAR monitoring with automatic TR extension when limits are approached.

8.7 Fat Suppression: Dixon vs. Spectral Methods

8.8 AI Reconstruction (Deep Learning Reconstruction, DLR)

Deep Learning Reconstruction (DLR) for TSE — implemented as Siemens Deep Resolve, GE AIR Recon DL, Philips SmartSpeed — applies convolutional neural network denoising to TSE raw data or image data post-acquisition. Published clinical studies report SNR gains equivalent to 2–4× NEX increase at equivalent scan time for brain and spine TSE [D4]. This enables either: (a) equivalent image quality at 50% scan time; (b) improved resolution at equivalent scan time. DLR does not alter the fundamental k-space acquisition and is retrospective — it cannot correct for through-k-space T2 modulation (blurring) caused by ETL. Caution is warranted regarding potential lesion smoothing at aggressive reconstruction settings.

9. Comparison with Alternative Sequences

When TSE is preferred:

TSE is the default T2-weighted sequence for virtually all clinical routine applications. It should be preferred when: adequate scan time permits multi-shot acquisition (>2 min); high in-plane resolution is required; fat suppression can be applied; field homogeneity is adequate. The clinical breadth of TSE is unmatched by any single alternative sequence.

When alternatives are superior:

Conventional SE remains the reference standard for quantitative T2 measurement and in musculoskeletal applications where fat J-coupling effect would confound interpretation. Single-shot TSE is superior in motion-prone environments. GRE is superior for T2*-sensitive applications (haemorrhage, calcification, SWI). EPI is mandatory for diffusion and perfusion. There is no universal "best" sequence — the value of TSE lies in its near-universal applicability combined with exceptional acquisition efficiency.

10. Evidence Gaps and Ongoing Debate

Optimal ETL for specific anatomical applications:

The selection of ETL for TSE T2 imaging is largely based on expert practice and technical convention rather than systematic comparative trials. The ETL range 8–24 for T2-weighted brain TSE represents accumulated expert consensus. Whether ETL=8 versus ETL=16 produces clinically meaningful differences in lesion detection for specific indications (e.g., MS plaque count, metastasis detection) is not definitively established in high-quality prospective comparative studies.

Variable flip angle TSE — contrast standardisation:

The VFA-TSE family (SPACE/CUBE/VISTA) introduces vendor-specific flip angle trains that produce tissue contrast that is not identical across platforms and not identical to standard TSE. The clinical implications of these contrast differences for specific indications — particularly in neuroradiology and musculoskeletal imaging — are not fully characterised. Comparative studies are predominantly single-centre technical papers. Protocol standardisation across institutions using different vendors remains an unresolved challenge.

AI reconstruction and diagnostic validity:

Deep Learning Reconstruction for TSE has been evaluated predominantly for SNR improvement and patient throughput. Whether DLR-TSE maintains diagnostic equivalence for specific detection tasks — particularly small lesion detection (MS plaques, inner ear pathology) — compared to conventional TSE at matched SNR is an active area of investigation. Available data are moderately encouraging but largely limited to retrospective technical validations [D4].

T1 TSE versus T1 GRE for post-contrast imaging:

In most protocols, post-contrast T1 imaging uses either TSE (short ETL) or 3D GRE (VIBE/LAVA/THRIVE). The superiority of one over the other for specific applications — particularly meningeal enhancement, small brain metastases, head and neck — is debated. 3D GRE offers volumetric isotropic acquisition; T1 TSE offers potentially higher T1 sensitivity at equivalent scan time in 2D. No definitive comparative consensus exists [Expert].

Fat suppression technique selection at 3T:

Dixon fat-water separation is widely considered superior to spectral fat saturation at 3T due to B0 inhomogeneity robustness, but this preference is based largely on expert consensus and single-centre comparisons. Head-to-head prospective multicentre studies for specific clinical indications are limited.

2D versus 3D TSE for spinal imaging:

High-resolution 3D TSE (SPACE/CUBE/VISTA) is technically capable of replacing 2D TSE in spinal protocols but introduces specific contrast differences and requires careful ETL and VFA optimisation. The clinical superiority of 3D over 2D for routine spine indications has not been demonstrated in adequately powered prospective trials.

11. Miscellaneous and Related Topics

PROPELLER/BLADE TSE:

Radial k-space TSE (PROPELLER/BLADE) acquires rotating strips of k-space, enabling retrospective motion correction from the oversampled k-space center. This is a clinically important TSE variant for motion-prone patients (paediatric, dementia, head trauma) in whom conventional TSE produces unacceptable motion artefact. Each "blade" is a short TSE echo train; multiple blades at different angles cover k-space. Oversampling at k-center provides motion correction data. The technique adds scan time relative to conventional TSE but substantially reduces motion sensitivity.

Hyperecho TSE:

Hyperecho is a Siemens-specific TSE implementation in which the flip angles of refocusing pulses are modulated to produce a coherent echo at the end of the echo train that is much larger than predicted by T2 decay alone. This reduces SAR substantially while maintaining central k-space signal. It is the precursor concept to modern VFA-TSE and is implemented on older Siemens platforms.

GRASE (Gradient and Spin Echo):

GRASE combines GRE readouts between SE refocusing pulses, filling multiple k-space lines per echo. This further increases acquisition speed beyond standard TSE but introduces mixed T2/T2* contrast and susceptibility sensitivity. GRASE was developed as an experimental acquisition strategy and has not achieved widespread clinical adoption. It represents a conceptual bridge between TSE and EPI.

Dark fluid TSE (dark blood MRI):

Application of double inversion recovery (DIR) or MSDE (motion-sensitised driven equilibrium) preparation to TSE produces dark-fluid or black-blood contrast, suppressing both CSF and blood simultaneously. Used in multiple sclerosis cortical lesion imaging (DIR-TSE) and cardiac MRI (MSDE-TSE). Child pages for these applications will develop these variants in detail.

Elastography-compatible TSE:

MR elastography (MRE) uses motion-encoding gradients to quantify tissue stiffness. Some MRE implementations use TSE-based readouts rather than GRE-based readouts to reduce susceptibility effects in certain anatomical regions.

Future directions:

Sub-millimetre 3D TSE for inner ear, spinal cord, and cranial nerve imaging continues to advance, driven by gradient hardware improvements and DLR denoising. Simultaneous multi-slice (SMS) TSE is under investigation to further accelerate 2D multislice acquisitions. Quantitative TSE — acquiring multiple TE points to generate pixel-wise T2 maps — is increasingly available on vendor platforms and represents a pathway from qualitative to quantitative tissue characterisation.

12. Bibliography

A. Guidelines / Consensus / Society Recommendations

B. Systematic Reviews / Meta-analyses

C. Important Prospective / Original Studies

D. Technical MRI Papers

E. Landmark Historical References

Document status: Active. Parent page: 9003 — MRI Sequences: Overview and Classification. Last evidence review: May 2026.