MR Angiography: TOF and Phase-Contrast QC

Introduction

Time-of-flight (TOF) and phase-contrast (PC) MR angiography produce vascular images from two entirely different physical mechanisms, so they require different acquisition choices and different quality-control checks. TOF is a bright-blood morphology technique driven by inflow of unsaturated spins; PC is a quantitative velocity map built from the phase shift that moving spins accumulate in a bipolar gradient. ^{1, 6}

Because non-contrast MRA avoids gadolinium, it has become a workhorse for intracranial, carotid, renal, and peripheral vascular imaging, and 4D flow has extended the phase-contrast idea into full three-dimensional, time-resolved velocity mapping of the heart and great vessels. ^{2, 6, 7} But the same physics that makes these sequences powerful also makes them fail in characteristic ways: TOF saturates slow and in-plane flow, and PC aliases flow faster than its velocity-encoding value. A medical physicist's job is to make sure the facility understands which mechanism each protocol relies on and verifies it with the right phantom test.

This guide explains how TOF and PC MRA work, derives the velocity-encoding (VENC) relationship that governs phase-contrast accuracy, and lays out a practical QC program that separates equipment performance from flow-specific behavior. DRPS provides this analysis as part of its MRI physics testing and accreditation support services across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware.

Topic Explanation

Two ways to make blood bright without contrast

Non-contrast MRA exploits the motion of blood itself rather than an injected agent. Two established mechanisms dominate clinical practice: ^{1, 3}

• Time-of-flight (inflow) enhancement. A thin slab or slice is excited rapidly and repeatedly. Stationary tissue inside the slab never fully recovers its longitudinal magnetization between pulses, so it becomes partially saturated and dark. Blood flowing into the slab carries fully relaxed, unsaturated spins that give a strong signal—so vessels appear bright against a suppressed background.
• Phase contrast (velocity-induced phase). A bipolar (flow-encoding) gradient is applied. Stationary spins accumulate zero net phase, but spins moving along the gradient accumulate a net phase proportional to their velocity. Subtracting two acquisitions with opposite gradient polarity isolates that velocity-dependent phase, yielding both a magnitude angiogram and a quantitative velocity map. ¹

A third family—contrast-enhanced MRA (CE-MRA) with gadolinium or, in renal-impaired patients, ferumoxytol—shortens blood T1 so vessels are bright regardless of flow, and is still widely used for large-field vascular studies. ⁸ This guide focuses on the two non-contrast, flow-dependent techniques, because they are the ones whose image quality is governed by flow physics and therefore need flow-specific QC.

Where each technique is used

TOF is the default for intracranial arterial imaging (circle of Willis) and is widely used for carotid and peripheral arteries, where its bright-blood morphology and high spatial resolution are the priority. ⁵ Phase contrast is used where quantification matters—cerebrospinal fluid flow, shunt fraction, valvular regurgitation, and velocity mapping—because it returns actual velocities and flow volumes, not just a picture. ¹ For background on how flow artifacts and gradient moment nulling are handled in routine imaging, see our companion article on MRI flow phenomena and flow compensation.

Key Technical Principles

The phase-contrast velocity equation

Phase contrast is the more quantitative of the two techniques, and its physics is worth making explicit. A bipolar gradient has a first gradient moment

A spin moving with velocity $v$ along the gradient direction accumulates a phase

where $\gamma$ is the gyromagnetic ratio. The velocity-encoding value, VENC, is defined as the velocity that produces a phase shift of exactly $\pi$ radians (180°):

Because phase can only be measured unambiguously between $-\pi$ and $+\pi$, any true velocity with magnitude greater than VENC aliases: the phase wraps, and a fast forward flow is displayed as reverse flow. ¹

Worked example. Suppose a carotid study is run with VENC = 150 cm/s and a voxel in the lumen returns a measured phase of 60°:

If the true peak systolic velocity were instead 180 cm/s—above the 150 cm/s VENC—the phase would wrap and the scanner would report an incorrect, aliased velocity. Raising VENC above the expected peak removes the wrap. The cost is noise: the velocity-to-noise ratio (VNR) is inversely related to VENC, so an unnecessarily high VENC makes slow flow noisy. Phantom work confirms this linear relationship between velocity noise and the chosen VENC. ^{3, 4} The clinical rule of thumb is to set VENC just above the expected peak velocity—close enough to preserve VNR, high enough to avoid aliasing.

What drives TOF signal

TOF has no single clean equation like VENC, but its behavior is governed by three levers:

• Repetition time (TR) and flip angle set how strongly stationary tissue is saturated. A shorter TR and higher flip angle darken the background more, but also saturate slow-moving blood that lingers in the slab.
• Slab or slice thickness and orientation determine how long inflowing spins stay unsaturated. Thin slices refreshed by through-plane flow stay bright; thick slabs or in-plane vessels saturate. Multiple overlapping thin slabs (MOTSA) and ramped (TONE) flip angles are used to keep distal vessels bright across a 3D slab.
• Saturation bands placed on one side suppress either arterial or venous inflow to separate arteries from veins.

The practical consequence is that TOF is exquisitely sensitive to flow geometry: slow flow, in-plane flow, and retrograde flow near stenoses can all drop out, mimicking disease. ⁵ That is a physics limitation, not an equipment fault, and a QC program must distinguish the two.

Comparing the techniques

The point of the table is operational: a protocol that is failing because of TOF saturation cannot be fixed by adjusting VENC, and an aliased phase-contrast study cannot be rescued by changing the flip angle. Matching the symptom to the mechanism is the first diagnostic step. ^{1, 2, 8}

Clinical Impact

When the physics masquerades as pathology

The most important clinical consequence of these mechanisms is that normal flow can imitate disease. On TOF, slow flow distal to a tight carotid or intracranial stenosis saturates and drops out, exaggerating the apparent degree of narrowing; intraluminal signal changes on TOF source images have even been studied as surrogate markers of upstream stenosis. ⁵ On phase contrast, an under-set VENC aliases peak systolic velocity and can turn a measured jet into a nonsensical reverse-flow reading, corrupting a regurgitant- or shunt-fraction calculation. ¹

Quantitative studies amplify the stakes. A shunt fraction (Qp/Qs), a valvular regurgitant fraction, or a 4D-flow–derived pressure gradient is only as good as the velocity measurement underneath it. The 4D flow consensus statement reports typical spatial resolution of roughly 1.5–3 mm, temporal resolution of about 30–40 ms, and acquisition times of about 5–25 minutes for whole-heart coverage—parameters that themselves affect velocity accuracy and must be held stable for serial comparisons. ⁶ If the velocity scale drifts or aliases, the derived clinical number drifts with it.

Why accreditation and serial consistency depend on QC

Vascular findings are frequently followed over time—aneurysm surveillance, stenosis progression, post-treatment flow. That makes reproducibility a clinical requirement, not just a nicety. A phase-contrast velocity that reads 10% high because of an eddy-current phase offset, or a TOF protocol whose background suppression has drifted after a gradient service, can change a management decision. Tying MRA performance into the facility's broader MRI program—see MRI acceptance testing and commissioning—keeps these measurements anchored.

Practical Optimization Tips

Phase contrast

• Set VENC just above the expected peak velocity. Start from published or prior-exam velocities, then confirm no aliasing on the first acquisition. If aliasing appears, raise VENC and repeat. ¹
• Correct for background phase offsets. Eddy currents and concomitant (Maxwell) gradient terms add a spatially varying phase that biases velocity. Use the vendor's background-phase correction and a static-tissue reference region; verify the correction on a stationary phantom. ^{1, 6}
• Keep the encoding direction right. Through-plane encoding for flow quantification; in-plane or three-directional encoding for flow visualization and 4D flow.
• Hold parameters constant for follow-up. For serial studies, lock VENC, resolution, and timing so the velocity measurement is comparable.

Time-of-flight

• Tune flip angle and TR to the vessel. Enough saturation to darken the background, not so much that slow flow drops out. Use ramped (TONE) flip angles across a 3D slab to keep distal vessels bright.
• Use thin slabs/MOTSA and saturation bands. Overlapping thin slabs reduce in-slab saturation; a saturation band removes venous (or arterial) overlap.
• Read the source images, not just the MIP. Maximum-intensity-projection reconstructions can hide or exaggerate stenoses; a dropout on the MIP should be confirmed on source images before it is called disease. ⁵

Both

• Minimize and recognize motion and pulsatility artifacts, and apply flow compensation where appropriate—see MRI flow phenomena and flow compensation.
• Keep geometric accuracy in check, because MRA vessel measurements inherit any gradient-driven spatial distortion; see MRI geometric distortion QC.

A practical MRA QC checklist

A defensible MRA QC program layers flow-specific tests on top of the routine MRI equipment program:

1. Baseline MRI equipment QC — SNR, geometric accuracy, slice thickness/position, and low-contrast performance on the ACR phantom, as in ACR MRI phantom QC.
2. Phase-contrast velocity accuracy — a flow phantom (constant-flow pump or rotating phantom) with a known velocity; confirm the measured velocity matches within tolerance and that aliasing appears at the predicted VENC.
3. Background phase-offset check — a stationary phantom should read zero velocity across the field; a nonzero reading flags eddy-current or concomitant-gradient bias.
4. TOF background suppression and saturation — verify stationary-tissue suppression and expected inflow behavior; confirm saturation bands work.
5. MIP fidelity — confirm reconstructions preserve vessel continuity without introducing or hiding stenoses.
6. Serial stability — trend the above so a post-service drift is caught before it reaches patients.

Regulatory Considerations

MR angiography is a non-ionizing modality, so its quality framework is MRI equipment performance and MR safety, not radiation dose. The governing documents are the ACR–AAPM technical standard for MRI equipment performance monitoring and the ACR MRI Accreditation Program, supported by the facility's MR safety program. ^{9, 10}

Key frameworks to reference:

• ACR–AAPM Technical Standard for Diagnostic Medical Physics Performance Monitoring of MRI Equipment (Revised 2019). Defines the qualified medical physicist's role in acceptance testing and annual performance monitoring of MRI systems, including the equipment parameters that underlie MRA image quality. ⁹
• ACR MRI Accreditation Program. Specifies phantom and clinical-image requirements and the qualified-physicist survey that accredited MRI sites must maintain. ¹⁰
• The Joint Commission diagnostic imaging requirements, which expect an annual MRI performance evaluation by a qualified medical physicist and an active MR safety program. ¹¹

Because MRA velocity quantification feeds clinical decisions, the physics evaluation should document not only standard equipment metrics but also velocity accuracy and background-phase behavior where phase-contrast or 4D flow is performed. Facilities building or renewing an MRI program should connect this to MRI physics testing, accreditation support, and medical physicist consulting. Note that none of this falls under NRC or Agreement State radioactive-material rules—MRI is regulated through accreditation and state facility requirements, not 10 CFR.

Frequently Asked Questions (FAQs)

What is the difference between TOF and phase-contrast MR angiography?

Time-of-flight (TOF) MRA creates bright-blood images from inflow enhancement: fresh, unsaturated spins flowing into a repeatedly excited slab are brighter than the suppressed stationary background. Phase-contrast (PC) MRA instead encodes velocity directly as a phase shift using bipolar gradients, so it produces both an angiogram and a quantitative velocity map. TOF shows vascular anatomy; PC measures how fast and in which direction blood is moving.

What is VENC in phase-contrast MRI?

VENC, the velocity-encoding value, is the velocity that produces a phase shift of 180 degrees (pi radians) in a phase-contrast sequence. It sets the maximum velocity the scan can measure without aliasing. A VENC set too low causes velocity aliasing (phase wrap); a VENC set too high wastes velocity-to-noise ratio and makes slow flow noisy. Choosing VENC just above the expected peak velocity is the central trade-off in phase-contrast imaging.

Why does flow look bright on TOF but can disappear on phase contrast?

TOF relies on unsaturated spins flowing into the imaging volume, so slow or in-plane flow that stays in the excited slab becomes saturated and loses signal. Phase contrast depends on the velocity relative to VENC, so flow faster than VENC aliases and flow much slower than VENC is buried in noise. The two techniques fail in different ways, which is why a QC program checks each separately.

What causes velocity aliasing and how is it corrected?

Velocity aliasing occurs when the true velocity exceeds the VENC, so the phase wraps past plus or minus 180 degrees and a high forward velocity is displayed as reverse flow. It is corrected by raising the VENC above the true peak velocity and re-acquiring, or by applying phase-unwrapping algorithms. On QC phantoms, aliasing is confirmed by comparing the measured velocity against the known pump or rotating-phantom velocity.

How is MR angiography quality controlled?

MRA QC combines the standard MRI equipment checks—signal-to-noise ratio, geometric accuracy, and slice performance—with flow-specific tests. A flow phantom with a known, adjustable velocity verifies that phase-contrast velocity measurements are accurate and that aliasing appears at the predicted VENC, while TOF acquisitions are checked for background suppression, saturation behavior, and maximum-intensity-projection fidelity. These checks support ACR accreditation and the facility's MRI physics program.

Does phase-contrast MRI involve ionizing radiation?

No. MR angiography, both time-of-flight and phase-contrast, uses magnetic fields and radiofrequency pulses, not ionizing radiation. The relevant safety and quality framework is MRI safety and MRI equipment performance—static field, gradients, RF heating (SAR), and acoustic noise—rather than radiation dose. Quality control therefore focuses on image fidelity, velocity accuracy, and MR safety screening.

What is 4D flow MRI and how does it relate to phase contrast?

Four-dimensional (4D) flow MRI is a time-resolved, three-dimensional phase-contrast acquisition that encodes velocity in all three directions over the cardiac cycle. It uses the same bipolar-gradient velocity-encoding physics as two-dimensional phase contrast, so the same VENC trade-offs and accuracy checks apply, but it allows retrospective flow measurement through any plane in the acquired volume.

Key Takeaways

• TOF and PC are different physics. TOF is inflow-driven bright-blood morphology; PC is velocity-encoded quantification. Each needs its own protocol choices and its own QC test.
• VENC is the master control for phase contrast. Because $\varphi =\pi v/\text{VENC}$, flow above VENC aliases and flow far below VENC is noisy. Set VENC just above the expected peak velocity.
• TOF fails by saturation. Slow, in-plane, and retrograde flow drop out and can mimic stenosis; read source images, not just the MIP.
• Quantification demands accuracy checks. Velocity-derived numbers (shunt fraction, regurgitant fraction, 4D-flow gradients) are only as good as the measured velocity, so phantom velocity accuracy and background-phase correction matter.
• MRA QC layers on top of MRI equipment QC. Standard ACR-phantom checks first, then flow-specific velocity and saturation tests.
• This is a non-ionizing modality. The governing framework is MRI performance and MR safety, not radiation dose.

Conclusion

MR angiography looks like a single capability on the scanner menu, but it is really two different physics experiments wearing the same label. Time-of-flight turns inflow into contrast and fails when flow is too slow or too parallel to the slab; phase contrast turns velocity into phase and fails when flow exceeds its velocity-encoding value. A facility that understands which mechanism a given protocol depends on—and verifies it with the matching phantom test—will read vascular studies with confidence and defend its quantitative numbers over time.

For the medical physicist, the deliverable is a QC program that separates equipment performance from flow-specific behavior: baseline ACR-phantom metrics, phase-contrast velocity accuracy against a known flow, background-phase verification on stationary tissue, and TOF saturation and MIP checks. Built that way, MRA stays accurate where it matters most—at the edge of a stenosis and in the middle of a velocity measurement.

How DRPS Can Help

Diagnostic Radiation Physics Services helps imaging facilities keep MR angiography accurate and accreditation-ready. This includes MRI physics testing and annual performance evaluations, flow-phantom velocity-accuracy checks for phase-contrast and 4D flow, TOF protocol review, accreditation support for the ACR MRI program, and medical physicist consulting for new sequences or scanner upgrades.

DRPS supports facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware.

Good MRA QC is not about running more sequences. It is about knowing which physical effect each image depends on—and proving, with a phantom, that the scanner is measuring it correctly.

Related Resources

• MRI flow phenomena and flow compensation
• ACR MRI phantom QC
• MRI geometric distortion QC
• MRI acceptance testing and commissioning
• Gadolinium contrast agents: relaxivity and safety
• MRI physics testing
• Accreditation support
• Medical physicist consulting

References

1. Wymer DT, Patel KP, Burke WF, Bhatia VK. Phase-Contrast MRI: Physics, Techniques, and Clinical Applications. Radiographics. 2020;40(1):122-140. doi:10.1148/rg.2020190039. doi.org
2. Miyazaki M, Akahane M. Non-contrast enhanced MR angiography: established techniques. J Magn Reson Imaging. 2012;35(1):1-19. doi:10.1002/jmri.22789. doi.org
3. Nilsson A, Markenroth Bloch K, Carlsson M, Heiberg E, Ståhlberg F. Variable velocity encoding in a three-dimensional, three-directional phase contrast sequence: evaluation in phantom and volunteers. J Magn Reson Imaging. 2012;36(6):1450-1459. doi:10.1002/jmri.23778. doi.org
4. Johnson KM, Markl M. Improved SNR in phase contrast velocimetry with five-point balanced flow encoding. Magn Reson Med. 2010;63(2):349-355. doi:10.1002/mrm.22202. doi.org
5. Osmanodja F, Scheitz JF, Fiebach JB, Ganeshan R, Villringer K. Can intracranial time-of-flight-MR angiography predict extracranial carotid artery stenosis? J Neurol. 2022;269(5):2743-2749. doi:10.1007/s00415-021-10876-0. doi.org
6. Dyverfeldt P, Bissell M, Barker AJ, et al. 4D flow cardiovascular magnetic resonance consensus statement. J Cardiovasc Magn Reson. 2015;17(1):72. doi:10.1186/s12968-015-0174-5. doi.org
7. Barger AV, Peters DC, Block WF, et al. Phase-contrast with interleaved undersampled projections. Magn Reson Med. 2000;43(4):503-509. doi:10.1002/(sici)1522-2594(200004)43:4<503::aid-mrm3>3.0.co;2-0. doi.org
8. Kocaoglu M, Pednekar A, Fleck RJ, Dillman JR. Cardiothoracic Magnetic Resonance Angiography. Curr Probl Diagn Radiol. 2024;53(1):154-165. doi:10.1067/j.cpradiol.2023.10.001. doi.org
9. American College of Radiology, American Association of Physicists in Medicine. ACR–AAPM Technical Standard for Diagnostic Medical Physics Performance Monitoring of Magnetic Resonance Imaging (MRI) Equipment. Revised 2019. acraccreditation.org
10. American College of Radiology. ACR Magnetic Resonance Imaging (MRI) Accreditation Program Requirements. acraccreditation.org
11. The Joint Commission. Diagnostic Imaging Requirements. jointcommission.org