Siemens PET Flow (FlowMotion) Explained

Siemens PET Flow technology, marketed as FlowMotion™, replaces traditional step-and-shoot PET acquisition with continuous bed motion, producing more uniform axial image quality, better quantitative accuracy, and a more efficient workflow.¹² Instead of stopping at a series of fixed bed positions, the patient table moves smoothly through the bore while events are acquired continuously across the entire axial field of view.

The physics that makes this matter is axial sensitivity. In any ring PET system, detector sensitivity peaks at the center of the axial field of view (AFOV) and falls toward the edges, so step-and-shoot imaging stitches together overlapping beds and leaves periodic sensitivity dips at the seams. Continuous bed motion removes those seams, flattens the effective sensitivity profile along the patient, and replaces discrete bed counting with a single continuous variable — table speed — that directly controls counts per axial position.³⁴

In this edition of the PhysicsPulse™ Series, we explain how continuous bed motion works, why it improves image uniformity and SUV consistency, what the published comparison studies actually show, and how technologists and physicists use table speed to optimize protocols. Diagnostic Radiation Physics Services (DRPS) supports PET/CT programs across Florida, Maryland, Virginia, Washington DC, California, and Nevada, where consistent quantitative performance is essential for both clinical reliability and accreditation.

Introduction

PET Flow changes the acquisition variable from "how many beds" to "how fast the table moves," and that single change is what flattens axial sensitivity and simplifies scan-range definition. Step-and-shoot acquisition has been the PET workhorse for decades, and it works — but it forces the protocol to be expressed in whole bed positions, each with its own overlap, and it imposes start-stop table motion on the patient. Continuous bed motion (CBM) reframes the same coincidence-counting problem as a continuously moving frame of reference.³

That reframing has three practical consequences. First, axial sensitivity becomes nearly uniform along the scan, instead of rising and falling with each bed seam.³⁴ Second, the scan range can be defined to the millimeter rather than rounded up to the next bed, which trims unnecessary overscan and the dose and time that come with it. Third, image quality is tuned by a continuous knob — table speed in millimeters per second — rather than by adding or dropping a discrete bed.¹²

None of this changes a facility's quantitative obligations. CBM and step-and-shoot are both subject to NEMA NU 2 performance measurement, ACR accreditation, AAPM acceptance and quality-control expectations, and the same NRC or Agreement State rules for the administered radiopharmaceutical.⁵⁶⁷⁸ This guide walks through the acquisition physics, a side-by-side comparison, the worked math behind axial sensitivity and table speed, the clinical evidence, optimization tips, and the regulatory context that a defensible PET Flow program needs.

Topic Explanation: What Is Siemens PET Flow Technology?

Siemens PET Flow (FlowMotion) is a continuous bed motion acquisition method in which the table moves through the scanner at a defined speed while PET data are collected continuously, rather than in discrete, stationary bed positions.¹²

To appreciate the difference, it helps to define both approaches:

• Step-and-shoot PET moves the patient table to a fixed bed position, stops, acquires data for a set time, then moves to the next position. Each bed position typically covers approximately 15–26 cm depending on scanner design, and adjacent positions must overlap to maintain uniform image quality.
• Continuous bed motion (FlowMotion) moves the table smoothly through the scanner while data are acquired along the entire axial field of view. Acquisition is defined by table speed, typically expressed in millimeters per second, instead of by a count of bed positions.¹²

The slower the table speed, the longer the effective acquisition time per region — and the higher the image quality. This continuous acquisition is supported by reconstruction algorithms that accurately map each detected coincidence event to its correct spatial location despite the ongoing table motion.³

Key terms used in this guide

• Axial field of view (AFOV) — the patient length the detector ring covers without table movement. In CBM, the AFOV is the "window" that sweeps along the patient.
• Axial sensitivity profile — how detection efficiency varies with axial position. For a single static bed it is roughly triangular, peaking at the ring center.
• Table speed (v) — the continuous acquisition variable in CBM, in mm/s. It sets how long any given slice spends inside the sensitive AFOV.
• Overscan — the extra acquisition beyond the clinical region needed to ramp sensitivity up and down at the ends of the scan; in step-and-shoot this is the bed overlap, in CBM it is the lead-in and lead-out travel.
• Standardized uptake value (SUV) — the semiquantitative metric most affected by axial sensitivity nonuniformity, because it depends on accurate, position-independent activity concentration.

Key Technical Principles

Uniform axial sensitivity

The core physics advantage of PET Flow is uniform axial sensitivity across the entire scan range. In a ring PET system, detector sensitivity is highest at the center of the axial field of view and falls off toward the edges. In step-and-shoot imaging this produces sensitivity dips at bed-position boundaries, which is why overlapping bed positions are required. Continuous bed motion removes these transitions entirely, so every slice along the patient is sampled with comparable sensitivity.³⁴

Key benefits include:

• Uniform axial sensitivity across the entire scan range
• Elimination of bed-transition artifacts and overlap inefficiency
• Improved consistency in standardized uptake value (SUV) measurements
• Better image quality in long scan regions such as skull-to-thigh oncology studies

Step-and-shoot vs. continuous bed motion at a glance

The table below summarizes the differences that matter most to physicists and technologists. The axial-sensitivity and overscan rows are the physics core; the workflow rows are where the clinical day-to-day difference shows up.

The percentages in the overscan row are illustrative of how overlap repeats at every junction in step-and-shoot; confirm exact overlap and ramp values against your scanner's configuration.

Worked math: axial sensitivity profiles

The single-bed axial sensitivity of a cylindrical PET ring is well approximated by a triangular profile. With the AFOV spanning $-L/2\le z\le L/2$ about the ring center, the relative sensitivity is:

$$s(z)\approx s_{\max }\left(1-\frac{2|z|}{L}\right),|z|\le \frac{L}{2}$$

Sensitivity is maximal at the ring center ($z=0$) and falls linearly to zero at the axial edges ($z=\pm L/2$). This is why a single static bed cannot uniformly image a region as long as its own AFOV: the ends are starved of counts. In step-and-shoot, beds are spaced by a pitch $p<L$ so that the rising edge of one bed overlaps the falling edge of the next. The composite sensitivity at axial position $z$ is the sum over beds $n$:

$$S_{\text{SS}}(z)=\sum _{n}s(z-np)$$

Because the triangles only partially overlap, $S_{\text{SS}}(z)$ has a periodic ripple — relatively flat where two beds overlap well, dipping where they meet at their low-sensitivity tails. Tightening the overlap (smaller $p$) reduces the ripple but adds beds, time, and overscan.

In continuous bed motion, the same triangular AFOV response is convolved with uniform table travel. As the table moves at constant speed $v$, each axial point passes through the entire sensitivity profile. Its total sensitivity is the integral of the single-bed response over the dwell, which for constant speed is proportional to the area under the triangle rather than its peak:

$$S_{\text{CBM}}(z)=\frac{1}{v}{\int }_{-\infty }^{\infty }s(z^{\prime })d{z}^{\prime }=\frac{1}{v}\cdot \frac{s_{\max }L}{2}$$

This result is independent of $z$ across the interior of the scan — the convolution of a finite kernel with a constant-velocity sweep produces a flat-topped (trapezoidal) profile, with linear ramps only over the first and last $L$ of travel (the lead-in and lead-out). That flat top is the uniform axial sensitivity that PET Flow advertises, expressed in physics rather than marketing.³⁴

Worked math: table speed, dwell time, and counts

Because acquisition is governed by table speed, the effective acquisition time per axial position follows directly from the AFOV and the speed. A point remains within the sensitive axial window for the time it takes the table to advance one AFOV length:

$$t_{\text{eff}}=\frac{L}{v}$$

For example, with an axial field of view $L$ and table speed $v=1.0\text{mm/s}$, an axial point dwells in the sensitive window for $t_{\text{eff}}=L[\text{mm}]/1.0\text{mm/s}$ seconds; halving the speed to $0.5\text{mm/s}$ doubles that dwell. (Use your scanner's actual AFOV; the relationship — not a specific number — is the point.)

Counts accumulate in proportion to dwell time, so for a fixed activity concentration the counts per axial position scale inversely with speed:

$$N\propto t_{\text{eff}}=\frac{L}{v}$$

and, in the Poisson-counting limit, image signal-to-noise scales with the square root of counts:

$$\text{SNR}\propto \sqrt{N}\propto \frac{1}{\sqrt{v}}$$

So halving the table speed roughly increases counts per position twofold and improves SNR by about $\sqrt{2}\approx 1.41$, at the cost of doubling scan time for that region. The total scan time for a range of axial length $\ell$ (plus ramps) is simply:

$$T_{\text{scan}}=\frac{\ell +L}{v}$$

where the extra $L$ accounts for the lead-in and lead-out travel needed to bring the flat-top sensitivity up and down at the ends. This is the central optimization relationship for PET Flow: speed sets both the noise and the scan duration, and the physicist's job is to pick the speed that meets the count-statistics target for the clinical task. Absolute count rates, of course, also depend on injected activity, uptake time, patient size, and scanner sensitivity, which is why protocol speeds must be validated on the installed system rather than copied between scanners.

Table speed as the acquisition variable

Because acquisition is governed by table speed rather than discrete positions, technologists tune image quality by selecting how fast the table travels:

• Slow speed (high image quality): oncology or low-count/low-dose studies
• Moderate speed: standard oncology protocols
• Faster speed: whole-body screening or follow-up studies

This lets a protocol be matched directly to the clinical indication and patient size. Understanding the relationship between table speed and effective acquisition time per region is the central skill for PET Flow protocol optimization. For the foundational timing physics that pairs with sensitivity, see our explainer on the technical advantage of Time-of-Flight (TOF) in PET imaging.

Detector technology integration

FlowMotion is implemented across Siemens Biograph PET/CT systems, including:

• Biograph mCT
• Biograph Horizon
• Biograph Vision
• Biograph Vision Quadra

These platforms use detector technologies such as:

• Lutetium oxyorthosilicate (LSO) scintillation crystals
• Silicon photomultiplier (SiPM) detectors (Vision-class systems)
• Time-of-Flight (TOF) capability

Pairing continuous bed motion with TOF and SiPM detectors improves timing resolution, spatial resolution, quantitative accuracy, and scan efficiency together.¹² The payoff of a high-sensitivity, long-AFOV digital platform is concrete: in an independent NEMA NU 2-2018 characterization, the SiPM-based Biograph Vision Quadra — with its 106 cm axial field of view — measured a system sensitivity of 83 cps/kBq at the shorter maximum-ring-difference setting and 176 cps/kBq using the full ring difference, with a coincidence time resolution near 228–230 ps.⁴ Higher intrinsic sensitivity directly raises the counts available at any given table speed, which is what lets CBM trade speed for noise so effectively.

Clinical Impact

PET Flow's uniform sensitivity translates directly into more reliable images and more trustworthy quantitative values — especially in larger patients and over long axial coverage.

What the comparison studies show

The clinical literature has tested CBM against step-and-shoot head-to-head, and the consistent finding is equivalence in core image quality at matched acquisition time, with an edge in end-plane uniformity and patient experience. In a randomized, double-blind crossover of 68 oncology patients scanned both ways in the same session, image quality, SUVs, and PET–CT alignment did not differ between techniques, while CBM produced significantly better end-plane image quality and patients clearly preferred the smoother acquisition.⁹ A phantom-and-patient study matching acquisition duration likewise found comparable contrast and visual scores, with CBM showing lower coefficient of variation at the end planes — though it noted slightly higher central-plane noise at the configurations tested.¹⁰ On a SiPM-based Biograph Vision, CBM delivered spatial resolution and image quality equivalent to step-and-shoot at the same acquisition time, with image noise rising as table speed increased — exactly what the $\text{SNR}\propto 1/\sqrt{v}$ relationship predicts.³ A controlled phantom study isolating respiratory-type motion found the two acquisition modes equivalent when net true counts and motion were matched.¹¹

The takeaway for a quality program: CBM is not a free lunch on noise, but at properly chosen table speeds it matches step-and-shoot image quality while improving end-plane performance and workflow — and the speed knob gives the physicist a continuous lever to hit count-statistics targets.

Better image quality in larger patients

Continuous motion maintains consistent sensitivity across the scan length, which is critical in larger and obese patients where count density is otherwise hardest to preserve. Advantages include:

• Improved signal-to-noise ratio (SNR)
• Reduced image-noise variability
• More reliable lesion detectability
• Improved quantitative consistency (SUV accuracy)

Quantitative accuracy and dose

PET Flow maintains accurate quantitative measurements while enabling scan and dose optimization. Because the count distribution is more uniform, the system can support:

• Improved SUV consistency across scan regions
• Lower injected activity while maintaining image quality
• Shorter scan time without compromising diagnostic performance

These capabilities support ALARA principles while preserving diagnostic reliability. Accurate SUVs also depend on correct radiopharmaceutical handling and timing — see PET imaging uptake time: why it matters for quantification and image quality and our overview of common isotopes in PET and radiopharmaceutical therapy.

Patient comfort and motion reduction

Smooth, continuous table travel — without the abrupt start-stop motion of step-and-shoot — tends to improve patient comfort, reduce anxiety, and improve scan completion rates.⁹ Reducing patient motion in turn reduces motion artifacts and further improves image quality.

Practical Tips for Technologists

To optimize PET Flow imaging:

1. Ensure the patient is properly centered in the scanner bore.
2. Select an appropriate table speed based on clinical indication and patient size.
3. Use slower table speeds when higher image quality is required (for example, low-count regions or larger patients).
4. Verify that scan coverage matches the physician's protocol requirements, defining the range to the clinical region rather than rounding to whole beds.
5. Monitor reconstructed images for uniform image quality across the entire scan range, paying attention to the lead-in and lead-out planes.

A few additional practices help physicists and lead technologists keep CBM protocols defensible:

• Validate speeds against count statistics, not habit. Because $N\propto L/v$, a speed copied from another scanner with a different AFOV or sensitivity will not deliver the same counts. Confirm noise and SUV recovery on your own system.
• Use variable speed where the patient warrants it. Slowing the table over high-attenuation regions (abdomen/pelvis) and speeding it over the lungs and legs concentrates counts where they are hardest to get, mirroring the way step-and-shoot users vary time per bed.
• Account for the ramps in scan time and dose budgeting. The lead-in and lead-out add roughly one AFOV of travel to every scan, per the $T_{\text{scan}}=(\ell +L)/v$ relationship.
• Re-baseline after any reconstruction or software change. Sensitivity uniformity and SUV recovery should be re-verified, since these are the very metrics CBM is marketed on.

Workflow efficiency is a direct benefit of removing discrete bed positions: there is no start-stop table motion, scan-time variability is reduced, patient setup and scan initiation are faster, and throughput improves.

Regulatory and Quality Considerations

Continuous bed motion does not change a facility's quantitative obligations — accreditation and quality-control requirements still apply to PET/CT performance. PET/CT scanners running FlowMotion are subject to the same oversight as conventional acquisitions:

• Accreditation: ACR Nuclear Medicine / PET accreditation requires documented physics performance monitoring, with quantitative testing aligned to NEMA NU 2 performance measurements and the ACR–AAPM technical standard for PET/CT.⁵⁶⁷
• Acceptance and QC: AAPM Task Group 126 and the IAEA PET/CT quality-control atlas provide the framework for acceptance testing and routine QC, including SUV accuracy and uniformity checks that are especially relevant when sensitivity uniformity is a selling point of the acquisition mode.⁸¹²
• Performance measurement standard: NEMA NU 2 defines how sensitivity, spatial resolution, noise-equivalent count rate, scatter fraction, and image quality are measured, and it is the common reference point for comparing scanners and validating that a CBM protocol meets the manufacturer's specifications.⁵ Because CBM is defined by table speed, your acceptance and annual testing should document the speeds used clinically and confirm that image quality and SUV recovery at those speeds satisfy the NU 2 image-quality and accuracy targets.
• Radioactive material use: Administration of PET radiopharmaceuticals is governed by NRC or Agreement State regulations under 10 CFR Part 35, and by state codes such as Florida Administrative Code 64E-5 for facilities in Florida.¹³

A note on jurisdiction: the PET radiopharmaceutical is byproduct material regulated under 10 CFR Part 35 (administered by the NRC directly in Washington DC, and by the Agreement State programs in Florida, Maryland, Virginia, California, and Nevada), while the CT subsystem of the PET/CT is a radiation-producing machine regulated by state radiation-control programs. A complete PET/CT quality program addresses both.

Because DRPS provides board-certified diagnostic medical physics services across Florida, Maryland, Virginia, Washington DC, California, and Nevada, we routinely verify that PET Flow protocols meet both the manufacturer's quantitative specifications and the applicable accreditation and state-regulatory requirements. For Florida-specific obligations, see Florida radiation safety requirements for imaging centers.

Frequently Asked Questions (FAQs)

What is the main advantage of PET Flow over step-and-shoot?

Continuous bed motion gives uniform axial sensitivity across the entire scan range, eliminating the overlap and sensitivity dips found at step-and-shoot bed-position edges. It also allows flexible scan length and table speed while maintaining quantitative SUV accuracy.

How is acquisition defined in PET Flow?

Acquisition is defined by table speed in millimeters per second rather than by discrete bed positions. A slower table speed lengthens the effective acquisition time per region and improves image quality and signal-to-noise ratio.

Does PET Flow affect SUV quantitative accuracy?

PET Flow improves SUV consistency by producing a more uniform count distribution along the axial field of view. Removing bed-edge sensitivity variation reduces position-dependent quantitative error across long scans such as skull-to-thigh oncology studies.

Which Siemens scanners support FlowMotion?

FlowMotion continuous bed motion is implemented across Siemens Biograph PET/CT platforms including the Biograph mCT, Biograph Horizon, Biograph Vision, and Biograph Vision Quadra. Vision-class systems pair it with SiPM detectors and time-of-flight.

Does continuous bed motion lower radiation dose?

PET Flow itself does not change the injected radiopharmaceutical activity, but its improved sensitivity uniformity supports ALARA by enabling lower injected activity or shorter scan time for the same image quality. Any activity reduction must follow your facility's clinical protocols.

Is continuous bed motion image quality really equivalent to step-and-shoot?

Published phantom and patient studies report that, at matched acquisition time, CBM delivers spatial resolution and image quality comparable to step-and-shoot, with better end-plane uniformity and patient comfort.³⁹¹⁰ Some studies note marginally higher noise at fast table speeds, consistent with the inverse relationship between counts and table speed, so speed selection still matters.

How do I set table speed for a PET Flow protocol?

Table speed is chosen to deliver the effective scan time per axial position your reconstruction and count statistics require, since counts per position scale as $L/v$ and SNR as $1/\sqrt{v}$. The physicist and lead technologist should validate speeds against NEMA NU 2 image-quality and SUV-recovery targets on the installed scanner before clinical release.

Key Takeaways

• Siemens PET Flow (FlowMotion) uses continuous bed motion — the table moves at a defined speed while data are acquired across the whole axial field of view, instead of stopping at discrete bed positions.
• Acquisition is set by table speed in mm/s; slower speeds increase effective acquisition time per region and improve image quality and SNR, with counts scaling as $L/v$ and SNR as $1/\sqrt{v}$.
• Uniform axial sensitivity is the central physics benefit: convolving the triangular single-bed sensitivity with constant-velocity travel produces a flat-topped profile, removing the bed-edge dips and overlap inefficiency of step-and-shoot PET.
• Quantitative SUV accuracy is preserved, and published head-to-head studies show image quality comparable to step-and-shoot at matched time, with better end-plane uniformity and patient comfort.³⁹¹⁰
• FlowMotion runs on Biograph mCT, Horizon, Vision, and Vision Quadra, and combines with LSO crystals, SiPM detectors, and Time-of-Flight on Vision-class systems.
• Accreditation and QC obligations are unchanged — ACR, NEMA NU 2, AAPM TG-126, and applicable NRC/state rules still govern PET/CT performance, and acceptance testing should document the clinical table speeds used.

How DRPS Can Help

Diagnostic Radiation Physics Services provides board-certified diagnostic medical physics support for PET/CT programs across Florida, Maryland, Virginia, Washington DC, California, and Nevada. We perform acceptance testing and annual surveys, verify SUV calibration and axial uniformity for FlowMotion protocols, validate clinical table speeds against count-statistics and NEMA NU 2 targets, support ACR PET accreditation, and align your quality-control program with NEMA NU 2, AAPM TG-126, and applicable NRC/Agreement State and state-specific regulations. If you are optimizing PET Flow protocols or preparing for accreditation, contact DRPS to ensure your quantitative performance is documented and compliant.

Conclusion

Siemens PET Flow technology is a significant advance in PET acquisition: by replacing step-and-shoot imaging with continuous bed motion, it delivers uniform axial sensitivity, more consistent SUV quantification, greater workflow efficiency, and improved patient comfort. The physics is straightforward — a constant-velocity sweep convolves the triangular single-bed sensitivity into a flat-topped profile, and table speed sets counts per position and therefore noise. Used well, with table speed matched to clinical indication and patient size and a sound QC program behind it, FlowMotion lets technologists and physicists produce reliable, accreditation-ready PET images while supporting ALARA.

For broader context on PET/CT physics and safety, see our guides on PET/CT shielding calculations and the Sweep: best practices for decontamination in nuclear medicine.

Related Resources

• Time-of-flight (TOF) PET imaging
• PET uptake time and quantification
• Common PET & RPT isotopes
• Florida radiation safety requirements
• PET/CT shielding calculations guide
• PET/CT and nuclear medicine physics

References

1. Siemens Healthineers. Biograph Vision PET/CT System Technical Documentation. siemens-healthineers.com
2. Siemens Healthineers. FlowMotion™ Technology Overview. siemens-healthineers.com
3. Kumamoto K, Sato H, Tsutsui Y, et al. Continuous bed motion in a silicon photomultiplier-based scanner provides equivalent spatial resolution and image quality in whole body PET images at similar acquisition times using the step-and-shoot method. J Nucl Med Technol. 2022;50(2):135-141. doi:10.2967/jnmt.121.263240. doi.org
4. Prenosil GA, Sari H, Fürstner M, et al. Performance characteristics of the Biograph Vision Quadra PET/CT system with a long axial field of view using the NEMA NU 2-2018 standard. J Nucl Med. 2022;63(3):476-484. doi:10.2967/jnumed.121.261972. doi.org
5. National Electrical Manufacturers Association. NU 2 — Performance Measurements of Positron Emission Tomographs (PET). Rosslyn, VA: NEMA. nema.org
6. American College of Radiology. ACR Nuclear Medicine and PET Accreditation Program Requirements. accreditationsupport.acr.org
7. American College of Radiology / American Association of Physicists in Medicine. ACR–AAPM Technical Standard for Diagnostic Medical Physics Performance Monitoring of PET/CT Imaging Equipment. acr.org
8. American Association of Physicists in Medicine. AAPM Task Group 126: PET/CT Acceptance Testing and Quality Control. aapm.org
9. Schatka I, Weiberg D, Reichelt S, et al. A randomized, double-blind, crossover comparison of novel continuous bed motion versus traditional bed position whole-body PET/CT imaging. Eur J Nucl Med Mol Imaging. 2016;43(4):711-717. doi:10.1007/s00259-015-3226-z. doi.org
10. Yamashita S, Yamamoto H, Nakaichi T, Yoneyama T, Yokoyama K. Comparison of image quality between step-and-shoot and continuous bed motion techniques in whole-body 18F-fluorodeoxyglucose positron emission tomography with the same acquisition duration. Ann Nucl Med. 2017;31(9):686-695. doi:10.1007/s12149-017-1200-5. doi.org
11. Owaki Y, Nakahara T, Shimizu T, et al. Effects of breathing motion on PET acquisitions: step and shoot versus continuous bed motion. Nucl Med Commun. 2018;39(7):665-671. doi:10.1097/MNM.0000000000000852. doi.org
12. International Atomic Energy Agency. PET/CT Atlas on Quality Control and Image Artifacts. IAEA Human Health Series No. 27. Vienna: IAEA; 2014. iaea.org
13. U.S. Nuclear Regulatory Commission. 10 CFR Part 35, Medical Use of Byproduct Material. nrc.gov
14. Cherry SR, Sorenson JA, Phelps ME. Physics in Nuclear Medicine, 4th Edition. elsevierhealth.com
15. Bushberg JT et al. The Essential Physics of Medical Imaging, 3rd Edition. Open Library