MRI Sequences and Pulse Sequences — Overview and Classification

MRIninja Knowledge Base | Master / Reference Page Version 1.0 — May 2026

1. Introduction and Purpose of This Page

Every MRI protocol on MRIninja is built from a defined set of sequences. Each sequence produces tissue contrast through a specific combination of radiofrequency pulses, gradient switching events, and signal sampling timing. Understanding what a sequence is — physically, technically, and clinically — is the prerequisite for understanding why it appears in a protocol, what it detects, and what it misses.

This page provides the classification framework and technical foundation for all MRI sequences referenced across the MRIninja knowledge base. It does not repeat the pulse sequence physics already covered in clinical MRI textbooks; instead, it provides the practical classification structure, clinical contrast logic, key parameters, vendor naming conventions, and inter-sequence decision rules that protocol designers need.

Each sequence family described here has, or will have, a dedicated child page on MRIninja covering its clinical applications, parameter optimisation, artefacts, and vendor-specific implementation. This master page provides the taxonomy and the cross-sequence relationships.

For the complementary parameter-level framework controlling TR, TE, TI, bandwidth, matrix, acceleration, reconstruction and physiological synchronisation, see MRI Parameters — Overview and Classification.

2. The Fundamental Framework: What a Pulse Sequence Does

An MRI pulse sequence is a precisely timed programme of radiofrequency (RF) pulses, gradient pulses, and data sampling windows that collectively determine three things simultaneously:

1. Signal source: which tissue magnetisation is being sampled (longitudinal vs transverse; equilibrium vs perturbed)
2. Contrast mechanism: which physical property of the tissue is being exploited to produce signal differences between tissues (T1 relaxation, T2 relaxation, T2* relaxation, proton density, diffusion, flow, chemical shift, magnetisation transfer, susceptibility)
3. Spatial encoding: how the signal is assigned to a location in space (Fourier encoding via frequency-encoding and phase-encoding gradients, or k-space trajectories for 3D and specialised acquisitions)

The radiologist and technologist must understand all three simultaneously to predict how changing a parameter affects image contrast, SNR, acquisition time, and artefact vulnerability.

3. Primary Classification: The Signal Readout Architecture

The most clinically useful primary classification of MRI sequences divides them by the readout architecture — the mechanism by which the MR signal is formed and sampled after the initial RF excitation. This determines the fundamental imaging speed, the type of contrast achievable, and the artefact profile.

3.1 Spin Echo (SE) and Its Derivatives

In a spin echo sequence, a 90° excitation pulse creates transverse magnetisation, and a subsequent 180° refocusing pulse reverses the dephasing effects of static field inhomogeneities, producing a true spin echo at a time TE after the 90° pulse. The 180° pulse eliminates T2* dephasing from B0 inhomogeneities but does not reverse true T2 decay. The signal amplitude at echo time TE therefore reflects the true T2 of the tissue. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Conventional Spin Echo (SE) Sequence.

The spin echo family provides the most reliable and artefact-resistant tissue contrast in clinical MRI, at the cost of longer acquisition times compared to gradient echo methods. The 180° refocusing pulse substantially reduces susceptibility artefacts from metallic implants, bone-air interfaces, and field inhomogeneities — making SE-based sequences the preference when susceptibility artefacts are clinically problematic (posterior fossa brain MRI, post-operative spine, ankle near metallic hardware). For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Gradient Echo (GRE/FLASH) Sequence.

Key variants of the SE family:

Conventional SE (CSE): single 90°–180° pulse pair; one echo per TR. Rarely used in modern clinical practice due to extreme time inefficiency.

Turbo Spin Echo (TSE) / Fast Spin Echo (FSE): the dominant clinical workhorse sequence family. Multiple 180° refocusing pulses are applied after a single 90° excitation, producing a train of echoes at successive echo times — the echo train. Each echo in the train is phase-encoded differently, filling multiple lines of k-space per TR interval. The number of echoes per TR is the echo train length (ETL) or turbo factor. Increasing ETL reduces acquisition time proportionally (a TSE with ETL=16 acquires 16 k-space lines per TR, reducing scan time to 1/16 of a comparable conventional SE). The price is T2 blurring: since echoes at the end of the train have been T2-weighted longer than early echoes, tissues with short T2 (cartilage, ligaments, tendons, cortical bone) are blurred when ETL is long. TSE contrast depends critically on the effective TE — the echo time assigned to the central k-space lines (which determine low-spatial-frequency contrast). TSE has displaced conventional SE for virtually all clinical T1, T2, and PD-weighted imaging. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Turbo Spin Echo (TSE/FSE) Sequence.

Single-Shot TSE (SSFSE / HASTE / FASE): the extreme case of TSE — all k-space lines for a single slice are acquired in a single TR after a single excitation. Acquisition time per slice is approximately 100–300 ms. Immune to respiratory motion because the entire k-space is acquired before motion occurs. Used for rapid abdominal screening, fetal MRI, diffusion preparation (e.g., navigated TSE DWI), and MRCP. Heavy T2 weighting and some T2 blurring are inherent because the latter echo train echoes contribute to image formation. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Spin Echo DWI / Non-EPI DWI Sequence.

Inversion Recovery TSE: a 180° inversion pulse is applied before the TSE readout. The inversion allows T1-based tissue signal suppression at the appropriate inversion time (TI). The tissue whose longitudinal magnetisation passes through zero at TI produces no signal — this is the null point. Three clinically important IR variants are used depending on the tissue to be suppressed:

• STIR (Short TI Inversion Recovery): TI chosen to null fat signal (~150 ms at 1.5T; ~200 ms at 3T). Produces robust T2/FLAIR-equivalent fat-suppressed images that are B0-independent. Used in MSK and body MRI for bone marrow and soft tissue oedema. Contraindicated after gadolinium injection (gadolinium shortens T1 of enhancing tissues, shifting their null point).
• FLAIR (Fluid-Attenuated Inversion Recovery): TI chosen to null CSF (~2200 ms at 1.5T; ~1700–1900 ms at 3T). Produces T2-weighted images without the CSF-bright background, dramatically improving periventricular and cortical lesion conspicuity. The dominant brain lesion detection sequence.
• DIR (Double Inversion Recovery): two sequential IR pulses null both white matter and CSF, leaving grey matter bright. Used for cortical lesion detection in MS and epilepsy.

Driven Equilibrium TSE (DRIVE / RESTORE / FRFSE): a final 90° pulse at the end of the TSE echo train tips remaining transverse magnetisation back to the longitudinal axis, restoring equilibrium faster and enabling shorter TR for equivalent contrast. Useful in myelography and joint fluid assessment where short TR T2-weighted images with bright fluid are desired.

3.2 Gradient Echo (GRE) and Its Derivatives

In a gradient echo sequence, the signal is formed by a gradient reversal (frequency-encoding gradient switching) rather than a 180° RF refocusing pulse. This means: (1) GRE sequences are significantly faster than SE sequences; (2) static field inhomogeneity effects (T2* dephasing) are NOT refocused — GRE signal is T2*-weighted, not T2-weighted; (3) susceptibility effects from iron, blood products, air-tissue interfaces, and metallic implants are amplified and exploited.

The flip angle in GRE sequences is typically < 90° (5°–80°), allowing rapid TR values without complete longitudinal recovery. This is the Ernst angle principle: the optimal flip angle for a given TR and T1 is the Ernst angle, below which reducing the flip angle further reduces the signal.

Key variants of the GRE family:

Spoiled GRE (SPGR / FLASH / T1-FFE / FFE): residual transverse magnetisation from previous TR periods is actively destroyed (spoiled) before the next excitation, either by gradient spoiling or RF phase cycling. The steady-state signal reflects the T1 of the tissue (with shorter T1 → brighter signal for a given TR/flip angle). Used for 3D T1-weighted brain imaging (MPRAGE is derived from this), dynamic contrast-enhanced (DCE) imaging, and time-of-flight (TOF) MRA.

Balanced Steady-State Free Precession (bSSFP / True FISP / FIESTA / b-FFE / BASG): all gradient moments are balanced within each TR, maintaining both longitudinal and transverse steady-state magnetisation. Produces high SNR with T2/T1-weighted contrast (bright fluid and fat, dark tissues with short T2/T1 ratio). Extremely fast and SNR-efficient. Used in cardiac imaging (ventricular assessment), small bowel imaging, and as the basis of virtual MR contrast agents. Banding artefacts from off-resonance are a specific limitation — multiple averages or frequency scouting are required in field-inhomogeneous regions. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Balanced Steady-State Free Precession (bSSFP) Sequence.

T2*-Weighted GRE (T2* GRE / FLASH at long TE): long TE GRE sequences amplify T2* effects, producing signal loss at sites of susceptibility — blood products, iron, calcification, metallic fragments. Used for microbleed detection, cavernoma, haemorrhagic transformation of infarcts, and as the basis of SWI.

Susceptibility Weighted Imaging (SWI): a multi-echo GRE sequence with specific phase image post-processing. SWI multiplies the magnitude image by a phase mask derived from the local phase image to amplify susceptibility contrast. Phase images encode the direction and magnitude of susceptibility — paramagnetic substances (deoxyhaemoglobin, haemosiderin, iron, contrast agent) produce positive phase shift; diamagnetic substances (calcium) produce negative phase shift. SWI is more sensitive for microbleeds and venous blood than T2*-GRE alone. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Susceptibility Weighted Imaging (SWI) Sequence.

Magnetisation-Prepared Rapid GRE (MPRAGE / IR-SPGR / BRAVO / TFE-IR): an inversion recovery preparation pulse (180°) is applied before a rapid spoiled GRE readout train. The inversion allows T1-based contrast (similar to IR-TSE) combined with the speed of GRE readout. The clinical result: 3D isotropic T1-weighted brain imaging at 1 mm isotropic in 5–8 minutes. MPRAGE is the standard sequence for brain morphometry, cortical thickness analysis, and the basis of the BTIP pre- and post-contrast T1 sequences in neuro-oncology. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page MPRAGE / 3D T1 Magnetisation-Prepared GRE Sequence.

3.3 Echo Planar Imaging (EPI)

Echo planar imaging acquires all k-space lines for a single slice in a single shot following one RF excitation, by rapidly oscillating the frequency-encoding gradient to create a train of gradient echoes while simultaneously stepping the phase-encoding gradient. Acquisition time per slice: approximately 50–100 ms. This speed is what makes EPI the basis of DWI, fMRI, DSC perfusion, and BOLD imaging. For sequence-level protocol optimisation, vendor terminology and artefact management, see the dedicated MRIninja page Echo Planar Imaging (EPI) Sequence.

EPI carries a specific artefact profile that is distinct from SE and GRE sequences:

• Geometric distortion from B0 inhomogeneity: EPI's slow phase-encoding accumulates large phase errors in regions of field inhomogeneity (air-tissue interfaces, susceptibility transitions). This distortion is most severe at 3T and in the posterior fossa, frontal sinuses, and adjacent to metallic implants.
• Chemical shift ghosting in the phase direction: fat chemical shift in the phase direction of EPI is 10× larger than the same effect in SE sequences at equivalent resolution, producing ghost images from fat. Fat saturation is mandatory for all clinical EPI.
• Nyquist ghost (N/2 ghost): a specific EPI artefact from phase inconsistency between odd and even echo lines, producing a ghost shifted by N/2 pixels in the phase direction.

SE-EPI (spin echo EPI): the EPI readout is preceded by a 90°–180° RF pulse pair, providing T2-weighted contrast with some T2* suppression. Used for DWI because the SE preparation refocuses the T2* dephasing from the diffusion sensitisation gradients.

GRE-EPI (gradient echo EPI): the EPI readout follows a simple GRE excitation. T2*-weighted contrast. Used for fMRI (BOLD), DSC perfusion, and cardiac EPI imaging.

3.4 Hybrid and Specialised Sequence Architectures

3D TSE / Variable Flip Angle TSE (SPACE / CUBE / VISTA / isoFSE): conventional TSE is inherently 2D (slice-selective). 3D TSE acquires a volumetric k-space by applying phase encoding in two directions after a non-selective or slab-selective excitation. A variable flip angle scheme along the echo train maintains the transverse magnetisation more efficiently, enabling long ETLs (100–300 echoes) that cover the 3D k-space volume in a single acquisition. The result is isotropic 3D volumes reformattable in any plane — critical for structural epilepsy MRI, inner ear imaging, vessel wall imaging, and MSK joint assessment. The variable flip angle is vendor-implemented differently: SPACE (Siemens), CUBE (GE), VISTA (Philips), isoFSE (Canon), with slightly different contrast and blurring profiles at equivalent resolution.

Parallel Imaging: not a pulse sequence family but an acceleration technique compatible with any sequence. Multiple receiver coil elements simultaneously sample different spatial frequencies; the redundant spatial information from different coil positions allows reconstruction of aliased under-sampled k-space data. GRAPPA (Siemens), ASSET (GE), SENSE (Philips), SPEEDER (Canon), and their extensions (CAIPIRINHA, controlled aliasing in parallel imaging) are the primary implementations. Acceleration factors (R) of 2–4 are routinely applied; higher R increases SNR noise amplification (g-factor penalty) and requires calibration reference data.

Compressed Sensing (CS): exploits the sparsity of MRI data in transform domains (wavelet, Fourier) to reconstruct images from fewer k-space samples than the Nyquist theorem requires, by enforcing sparsity in the reconstruction. CS can be combined with parallel imaging for further acceleration (CS-SENSE, XD-GRASP). CS is increasingly used for dynamic contrast imaging, 4D flow, and accelerated MSK sequences.

Deep Learning Reconstruction (DLR): vendor-implemented neural network reconstructions (AIR Recon DL — GE; Deep Resolve — Siemens; SmartSpeed — Philips) that denoise undersampled acquisitions to produce images with SNR equivalent to longer acquisitions. DLR effectively allows higher acceleration factors than CS alone while maintaining diagnostic image quality. Clinically validated for brain, spine, knee, and cardiac imaging.

4. Contrast Classification: What Physical Property Is Being Imaged

Independent of the readout architecture, sequences can be classified by the primary contrast mechanism they exploit. This determines what the sequence shows, what pathology it detects, and what it misses.

4.1 T1-Weighted Contrast

T1 contrast reflects the rate of longitudinal magnetisation recovery after RF excitation (T1 relaxation time). Tissues with short T1 (fat, gadolinium-enhanced tissue, subacute haemorrhage, proteinaceous fluid, melanin, high protein concentration) recover quickly and appear bright. Tissues with long T1 (CSF, free water, tumour) recover slowly and appear dark.

Sequences producing T1 contrast: spoiled GRE (short TR, Ernst angle flip), TSE with short TR (400–700 ms), MPRAGE/BRAVO/TFE-IR, inversion recovery sequences at appropriate TI to produce T1-like discrimination.

Primary clinical uses: anatomy, gadolinium enhancement, fat characterisation, haemorrhage staging, white matter vs grey matter contrast in the brain.

4.2 T2-Weighted Contrast

T2 contrast reflects the rate of transverse magnetisation decay due to spin-spin interactions (true T2 relaxation, not T2*). Tissues with long T2 (CSF, free water, oedema, most pathology) appear bright. Tissues with short T2 (cortical bone, fibrous tissue, tendons, ligaments, calcification) appear dark and signal-poor.

Sequences producing T2 contrast: TSE with long TE (60–120 ms), SE with long TE, single-shot TSE, 3D TSE (SPACE/CUBE/VISTA) with appropriate TE. Not achievable with GRE (which produces T2*, not T2, contrast).

Primary clinical uses: lesion detection (most pathology has elevated T2), fluid characterisation, internal organ morphology, joint fluid.

4.3 T2*-Weighted Contrast (Susceptibility Contrast)

T2* reflects combined decay from true T2 and additional dephasing from local field inhomogeneities (B0 heterogeneity from susceptibility differences between tissues). T2* is always shorter than or equal to T2. The relationship: 1/T2* = 1/T2 + γ·ΔB0 (where γ·ΔB0 is the local field inhomogeneity term).

Sequences producing T2* contrast: GRE sequences at long TE, SWI, EPI (inherently T2*-weighted without refocusing).

Clinical uses: detection of blood products (haemosiderin, deoxyhaemoglobin, methaemoglobin in older stages), iron deposition, calcification (though calcium can appear paradoxically bright or dark depending on the local field effect), microbleeds, cavernoma, venous anatomy.

4.4 Proton Density (PD) Weighted Contrast

PD contrast minimises both T1 and T2 contrast: TR is long enough for near-complete T1 recovery (eliminating T1 weighting) and TE is short enough to minimise T2 decay (eliminating T2 weighting). The residual signal differences reflect the concentration of mobile protons (proton density) in each tissue.

Sequences: TSE with long TR (≥ 2500 ms) and short TE (20–40 ms), with or without fat suppression. Standard for MSK imaging (rotator cuff, menisci, ligaments) where intermediate contrast between tendons and fluid is required.

4.5 Diffusion-Weighted Contrast

DWI exploits the restriction of water molecular diffusion by cellular and structural barriers. Diffusion-sensitising gradient pulses (Stejskal-Tanner bipolar gradients) are applied before and after the refocusing pulse in a SE-EPI sequence. Tissue with restricted diffusion (acute ischaemia, hypercellular tumour, abscess) retains high signal at high b-value; tissue with free diffusion loses signal proportionally to the b-value and ADC.

The apparent diffusion coefficient (ADC) map is calculated from at least two b-values: ADC = -[ln(S₁/S₀)] / (b₁ - b₀). True diffusion restriction is confirmed by DWI hyperintensity accompanied by ADC hypointensity (ADC < 700–800 × 10⁻⁶ mm²/s in acute ischaemia); T2 shine-through produces DWI hyperintensity with elevated (not reduced) ADC.

4.6 Perfusion-Weighted Contrast

Perfusion MRI measures cerebral haemodynamics. Two principal approaches:

DSC (Dynamic Susceptibility Contrast): T2*-EPI time series during an intravenous gadolinium bolus. The gadolinium bolus produces transient T2* signal loss (susceptibility effect) proportional to local blood volume. Post-processing deconvolves the time-signal curve with the arterial input function to produce CBF, CBV, MTT, and Tmax maps.

ASL (Arterial Spin Labelling): endogenous perfusion imaging without contrast. Water protons in inflowing arterial blood are magnetically labelled by RF inversion; the difference signal between labelled and unlabelled (control) images reflects the delivery of labelled water to tissue, providing a quantitative CBF map. Technical variants: pulsed ASL (PASL), continuous ASL (CASL), pseudo-continuous ASL (pCASL — the current standard).

4.7 Flow and Vascular Contrast

Time-of-Flight (TOF) MRA: flowing blood continuously refreshes fully relaxed magnetisation into the imaging slice, producing high signal against the stationary saturated tissue background. No contrast agent required. The saturation of in-plane flow limits TOF to predominantly perpendicular vessel assessment; long or tortuous vessels parallel to the imaging plane are saturated and may appear occluded.

Phase Contrast (PC) MRA / MRV: velocity-encoding gradients produce phase shifts proportional to spin velocity. The phase image encodes flow velocity; the magnitude image provides anatomical localisation. PC-MRA provides quantitative flow velocity measurement and directional flow information — the basis of 4D flow MRI.

Contrast-Enhanced MRA (CE-MRA): T1-spoiled GRE sequence during intravenous gadolinium injection. The gadolinium bolus shortens blood T1 below that of background tissue, producing selective vessel enhancement. CE-MRA provides large field coverage, freedom from TOF saturation artefacts in tortuous vessels, and excellent SNR at the cost of requiring contrast.

5. Classification by Dimensional Architecture: 2D vs 3D

5.1 2D Sequences

In 2D acquisition, each slice is independently excited and its k-space filled before moving to the next slice. The primary advantages are: (i) independence of slice k-space — motion between slices does not affect adjacent slice data; (ii) shorter TE achievable for thin slices; (iii) insensitivity to global k-space motion artefacts.

The primary limitations are: (i) inter-slice gaps and slice-selective excitation profile imperfections (cross-talk); (ii) no reformatting capability — the slice plane must be correct at acquisition time; (iii) SNR per unit time is lower than an equivalent 3D acquisition because the thin slice profile means less tissue per voxel.

2D sequences remain preferable for: DWI (EPI is inherently 2D); standard clinical T2 and T1 imaging where reformatting is not required; FLAIR; all sequences where motion between slices is a primary concern; and dynamic sequences (DCE, DSC).

5.2 3D Sequences

In 3D acquisition, the entire volume is excited as a slab (or the whole brain/organ), and phase encoding is applied in two orthogonal directions, filling a 3D k-space. Advantages: (i) isotropic voxels and MPR capability — the dataset can be reformatted in any plane after acquisition; (ii) higher SNR per unit time than equivalent 2D; (iii) no inter-slice gaps; (iv) enables advanced post-processing (morphometry, tractography, surface reconstruction). Limitations: (i) any motion during the acquisition affects the entire 3D volume (not isolated to one slice); (ii) wrapping in the slab direction; (iii) longer acquisitions; (iv) SAR constraints for large slab TSE acquisitions at 3T.

Clinical preference for 3D: MPRAGE/BRAVO (T1 morphometry), 3D FLAIR (MS and epilepsy), 3D TSE (isotropic joint assessment, vessel wall imaging), TOF MRA, 3D T1 post-contrast (oncology), DTI.

6. Classification by Clinical Contrast Target: The Practical Protocol View

From the clinical protocol design perspective, sequences are most usefully classified by their primary clinical contrast target — what they are intended to show in practice:

7. Vendor Naming Conventions — Cross-Reference Table

One of the most common sources of confusion in inter-institutional protocol communication is vendor-specific sequence names. The following table provides the cross-vendor equivalence for clinically important sequences across the four major manufacturers.

8. Key Acquisition Parameters and Their Roles

Every sequence is defined by a set of parameters that jointly determine image contrast, spatial resolution, SNR, and acquisition time. The following are the clinically essential parameters that protocol designers must understand.

8.1 TR (Repetition Time)

TR is the interval between successive RF excitations of the same slice or k-space position. TR primarily controls T1 weighting: short TR (300–700 ms) → T1-weighted (tissues recover differentially); long TR (≥ 2500 ms) → T1 contrast is minimised, and T2 or PD contrast dominates. At 3T, tissues have longer T1 than at 1.5T, requiring slightly longer TR for equivalent T1 weighting.

8.2 TE (Echo Time)

TE is the interval between the RF excitation and the centre of the signal readout (the echo peak). TE primarily controls T2 or T2* weighting: short TE (< 30 ms) → minimal T2/T2* decay → PD or T1-like contrast; long TE (60–120 ms) → T2-weighted; very long TE (> 150 ms) → heavily T2-weighted (free fluid bright, all solid tissue dark).

8.3 TI (Inversion Time)

TI is the interval between the inversion recovery 180° pulse and the subsequent excitation pulse (in IR sequences). TI controls T1-based signal suppression: the tissue whose longitudinal magnetisation passes through zero at TI is nulled. Different TI values null different tissues: short TI (150–200 ms) nulls fat (STIR); long TI (1700–2400 ms) nulls CSF (FLAIR).

8.4 ETL / Turbo Factor / Echo Train Length

The ETL (in TSE) determines how many k-space lines are acquired per TR. Higher ETL → shorter scan time; but also → increased T2 blurring (from the later echo contributions to k-space). The optimal ETL balances speed against acceptable blurring for the target structure (e.g., short ETL for cartilage; long ETL for FLAIR/T2 brain).

8.5 Flip Angle (FA)

FA is the tip angle of the longitudinal magnetisation produced by the RF excitation pulse. In SE sequences, the excitation FA is fixed at 90°. In GRE sequences, FA is a free parameter: low FA (5–20°) → PD-like; moderate FA (30–60°) → optimal T1 (near Ernst angle for typical tissues); high FA (60–90°) → T1-weighted. In TSE sequences with variable FA trains (SPACE/CUBE/VISTA), the FA of each refocusing pulse in the echo train is varied to maintain steady-state magnetisation over the long ETL.

8.6 b-value (in DWI)

b-value is the diffusion sensitisation factor, proportional to the area and separation of the diffusion gradient pulses (b = γ²G²δ²(Δ-δ/3) in standard Stejskal-Tanner notation). Higher b → greater diffusion weighting; at high b-values, only severely restricted tissue retains signal. Standard clinical b-values: b = 1000 s/mm² for brain DWI; b = 800–1000 for body; b ≥ 1400 for prostate (per PI-RADS).

8.7 Venc (Velocity Encoding) in Phase Contrast MRA

Venc determines the range of velocities that can be unambiguously measured. Velocities exceeding Venc produce phase aliasing (velocity mirroring). Venc must be set above the maximum expected velocity (arteries ~100–200 cm/s; veins ~20–40 cm/s) while being low enough for adequate sensitivity.

9. Acquisition Strategies That Modify All Sequences

Several technical strategies modify the fundamental behaviour of any sequence type described above. These are not sequences themselves but acquisition modifiers that protocol designers must be aware of:

Parallel imaging (described in Section 3.4): reduces acquisition time at the cost of SNR reduction proportional to √R for each acceleration direction.

Partial Fourier acquisition: acquires only slightly more than half of k-space, reconstructing the missing lines by conjugate symmetry. Reduces acquisition time; introduces minor phase errors and Gibbs ringing at very short ETL; reduces SNR modestly.

Fat suppression techniques: STIR, CHESS/ChemSat, SPAIR, Dixon, and water-selective excitation (Water Excitation) are all fat suppression strategies that can be applied as modifiers to almost any sequence. Their specific selection depends on field strength, B0 homogeneity, body region, and whether contrast has been administered — detailed in the Fat Suppression child page on MRIninja.

SMS (Simultaneous Multi-Slice / Multiband EPI): applies CAIPIRINHA or blipped-controlled aliasing pulses to simultaneously excite and encode multiple slices in a single TR, then unfolds them using parallel imaging. Substantially reduces total acquisition time for multi-slice EPI (DWI, fMRI, DSC perfusion) without changing TE or diffusion parameters.

Motion correction strategies: prospective motion correction (navigator-based for cardiac and body MRI; optical tracking for brain MRI); retrospective motion correction (PROPELLER/BLADE radial k-space trajectories for brain DWI and T2); and motion-robust single-shot acquisitions (HASTE, SS-EPI).

Compressed sensing / Deep learning reconstruction: described in Section 3.4. Applied as post-acquisition signal processing steps after any undersampled acquisition.

10. The Sequence Selection Decision Framework

Protocol designers face the practical question of how to choose between sequence options. The following framework provides a logical structure for this decision:

Step 1: Define the contrast target

What tissue property must be imaged? T1 (anatomy, gadolinium)? T2 (pathology, fluid)? Diffusion? Susceptibility? Flow?

Step 2: Define the structural requirements

What is the target structure size? A 1 mm cortical lesion requires 3D isotropic acquisition; a full-thickness meniscal tear requires thin 2D slices in the correct plane; a renal mass requires large FOV coverage.

Step 3: Define the environment constraints

Field strength (1.5T vs 3T) affects T1 values, SAR, B0 homogeneity, SNR, and susceptibility effects. Metal proximity requires SE over GRE. Off-isocentre positioning requires B0-independent fat suppression.

Step 4: Define the time constraint

Total acceptable acquisition time determines ETL, parallel imaging factor, 2D vs 3D choice, and whether full or abbreviated protocol is appropriate.

Step 5: Define the clinical decision outcome

What answer must the sequence provide? "Is there diffusion restriction in this lesion?" → DWI. "Is there gadolinium enhancement?" → T1-SE post-contrast. "What is the calibre of the MCA?" → TOF MRA. The clinical question drives the sequence choice.

11. Future Child Pages Building on This Framework

This master page provides the taxonomy and physical foundation for the following dedicated child pages on MRIninja, each covering one sequence family in full clinical and technical depth:

• Spin Echo and TSE sequences: ETL optimisation, T2 blurring, MSK applications, body T2
• STIR: musculoskeletal bone marrow, paediatric, post-contrast contraindication
• FLAIR and 3D FLAIR: brain lesion detection, MS protocol, epilepsy, SAH sensitivity
• MPRAGE and 3D T1-GRE: brain morphometry, BTIP T1 protocol, DCE
• SWI: haemorrhage, microbleeds, venous anatomy, CVS in MS
• DWI and ADC: acute stroke, oncology body DWI, prostate DWI, testicular DWI
• DSC perfusion: glioma grading, RANO response, penumbra assessment
• ASL perfusion: paediatric, epilepsy, late-window stroke
• TOF MRA: circle of Willis, extracranial cervical vessels, peripheral arteries
• Phase contrast MRA and 4D flow: venous sinus assessment, IIH, cardiac
• CE-MRA: aorta, renal arteries, peripheral runoff
• bSSFP / TrueFISP: cardiac, bowel, fetal MRI
• Vessel wall MRI (3D T1 black-blood): RCVS vs vasculitis, atherosclerosis, dissection
• DTI and tractography: white matter anatomy, surgical planning
• MR spectroscopy (MRS): brain metabolite profiles, glioma, hepatic steatosis
• Fat suppression techniques: Dixon, SPAIR, STIR, CHESS — selection guide
• 3D TSE (SPACE/CUBE/VISTA): joint cartilage, inner ear, epilepsy, vessel wall

New sequence child page: Inversion Recovery (IR) provides the parent physics of STIR, FLAIR and DIR, including TI selection, tissue nulling and field-strength effects.

12. Evidence-Based References

A. Guidelines / Consensus / Society Recommendations

(No society guidelines are dedicated to sequence classification per se; the evidence base for this foundational page rests on landmark technical papers and authoritative textbooks.)

B. Systematic Reviews / Meta-analyses

(Not applicable for this classification/foundational page.)

C. Important Prospective / Original Studies

(Not applicable for this foundational page.)

D. Technical MRI Papers

E. Landmark Historical References

End of document — MRI Sequences and Pulse Sequences: Overview and Classification — MRIninja v1.0 — May 2026 This master page is the reference for all future MRIninja child pages dedicated to individual sequence families.