Gamma Camera Uniformity QC: Floods and Limits

Gamma camera flood-field uniformity QC confirms that a detector counts a uniform radiation field evenly across its face, using the NEMA integral and differential uniformity metrics to catch photomultiplier drift, energy-calibration errors, and collimator defects before they become artifacts in a patient study. It is the single most performed test in a nuclear medicine QC program, and for good reason: nearly every other imaging problem eventually shows up as a non-uniformity.¹³

A gamma camera image is built from millions of individual scintillation events, and the camera must place each event at the correct location and assign it the correct energy. Uniformity testing floods the detector with a spatially uniform source and asks a simple question: does the camera record the same count density everywhere it should? When the answer drifts, a flood field is where you see it first.¹³

Introduction

Uniformity is the daily heartbeat of gamma camera quality control. Before the first patient, a technologist acquires a flood image and checks it — visually and numerically — against established limits. The test is quick, but the physics behind it, and the discipline of interpreting it correctly, are what make it valuable. A flood that "looks fine" can still hide a differential non-uniformity that becomes a ring artifact after SPECT reconstruction.¹³⁴

This guide explains what uniformity actually measures, defines the NEMA integral and differential uniformity metrics with worked calculations, distinguishes intrinsic from extrinsic testing, shows how counting statistics dictate how many counts a flood must contain, and connects daily QC to acceptance testing and accreditation. DRPS supports uniformity and full-system testing as part of its PET/CT and nuclear medicine physics and accreditation support services. For the acceptance-level performance framework this daily test sits within, see our guide to gamma camera NEMA NU-1 performance testing.

Topic Explanation

What is flood-field uniformity?

Flood-field uniformity describes how evenly a gamma camera records counts when its detector is exposed to a spatially uniform radiation field. An ideal detector exposed to a uniform flux would produce the same count density in every pixel; a real detector shows variations arising from several sources:

• Photomultiplier tube (PMT) response — differences in gain across the array of PMTs behind the crystal.
• Energy calibration — errors in the energy correction that shift how events near the photopeak are accepted.
• Spatial linearity — small position-encoding errors that pile up or thin out counts locally.
• Collimator defects — damaged or cast-defect septa that create localized count changes, visible only when the collimator is in place.
• Crystal defects and hydration — cracks or moisture ingress in the NaI(Tl) crystal.

Uniformity QC exists to quantify these variations and to catch a developing problem while it is still small.¹³ Zanzonico's review of routine nuclear medicine QC places flood-field uniformity among the core, frequently repeated instrument checks precisely because it is sensitive to so many failure modes.³

Intrinsic versus extrinsic uniformity

There are two ways to flood the detector, and each answers a different question:

• Intrinsic uniformity is measured with the collimator removed, using a point source of Tc-99m placed at a distance of at least five useful-field-of-view dimensions so the flux is effectively uniform. It isolates the detector, electronics, and energy correction.
• Extrinsic (system) uniformity is measured with the collimator in place, using a Co-57 sheet source (122 keV) or a fillable flood source of Tc-99m. It tests the entire imaging chain, including the collimator.

Both are needed. A collimator defect will not appear in an intrinsic flood, and some electronic problems are easiest to isolate with the collimator off. Many facilities perform daily extrinsic floods on the mounted collimator for convenience and periodic intrinsic floods for detector characterization.³⁶ The choice of source and geometry directly affects the result, which is one reason acceptance testing documents the exact method used.

Key Technical Principles

The NEMA uniformity metrics

The National Electrical Manufacturers Association standard NEMA NU 1-2018 defines how uniformity is calculated so that values can be compared across systems and over time.¹ Two quantities are reported, each evaluated twice — once over the useful field of view (UFOV) and once over the central field of view (CFOV), which is the central 75% of the UFOV linear dimensions.

Integral uniformity (IU) captures the worst overall variation:

where $C_{\max }$ and $C_{\min }$ are the maximum and minimum pixel counts within the region being evaluated, after the standard smoothing NEMA prescribes.

Differential uniformity (DU) captures sharp local gradients by taking the largest value of the same expression over any set of five contiguous pixels along each row and column:

Both are expressed as percentages, and lower is better. Integral uniformity can be small while differential uniformity flags a steep local change — exactly the kind of gradient that becomes a SPECT ring artifact — which is why both must be read together.¹⁷

Worked example: integral uniformity

Suppose a smoothed flood over the UFOV has a maximum pixel value of $C_{\max }=1050$ counts and a minimum of $C_{\min }=950$ counts. The integral uniformity is:

A 100-count spread on a roughly 1000-count background is a 5% integral non-uniformity. Modern large-field cameras are typically specified well below this value intrinsically, so a 5% reading would prompt investigation. Whether it represents a real defect or simply statistical noise depends entirely on how many counts the flood contains — which brings in counting statistics.¹⁷

Counting statistics set the required count density

Radioactive decay is a Poisson process, so a pixel that has recorded $N$ counts has a statistical standard deviation of $\sqrt{N}$, and a relative statistical fluctuation of:

To keep statistical noise from masquerading as non-uniformity, the per-pixel fluctuation must be small compared with the uniformity limit being tested. Requiring roughly a 1% statistical fluctuation per pixel means:

That per-pixel requirement drives the total count target. A daily flood stored in a modest matrix may therefore need several million counts, and a high-count uniformity-correction flood used for SPECT can require tens of millions of counts so that the correction map itself is not noisy.¹³ NEMA also specifies the acquisition matrix so that pixels are approximately 6.4 mm on a side, standardizing the spatial scale of the measurement.¹⁷

Uniformity metric summary

The specific numeric pass/fail limits belong to the manufacturer's specification and the facility's baseline, not to a universal number; the metrics above are the framework within which those limits are applied.¹⁶

Clinical Impact

Uniformity errors do not stay confined to the flood image — they propagate into patient studies, and SPECT amplifies them. In planar imaging, a modest non-uniformity may be tolerable or visually obvious. In SPECT, the reconstruction combines many angular projections through the same detector region, and a small, fixed differential non-uniformity is turned into a concentric ring artifact centered on the axis of rotation. A ring can mimic or obscure real pathology, so SPECT places tighter demands on uniformity than planar imaging alone.¹³⁴

This sensitivity is why nuclear medicine departments acquire dedicated high-count uniformity-correction floods and apply the resulting correction map during acquisition. It is also why differential uniformity — the metric most closely tied to sharp local gradients — deserves as much attention as the headline integral number. A recent multi-site survey of gamma camera and SPECT/CT QC programs found that vendor-based flood-field uniformity verification was performed universally, while more labor-intensive tests varied widely between sites, underscoring how central uniformity is to routine practice.⁴

Uniformity also interacts with the rest of the QC chain. A center-of-rotation error, an energy-peaking drift, or a collimator problem can each present first as a uniformity change, so a uniformity failure is often the start of an investigation rather than the end of one. For the companion tomographic checks, see SPECT center of rotation QC and SPECT/CT quality control.

Practical Optimization Tips

1. Peak the energy window before the flood

Uniformity is measured within the photopeak energy window, so a mis-peaked window will distort the result. Confirm the correct photopeak (140 keV for Tc-99m, 122 keV for Co-57) and window width before acquiring the flood, so that a peaking error is not misread as a uniformity failure.³⁶

2. Match the source and geometry to the test

For intrinsic floods, use a point source at an adequate distance (at least five UFOV dimensions) and a low enough count rate to avoid dead-time distortion. For extrinsic floods, use a well-maintained Co-57 sheet source or a properly mixed fillable flood, and confirm the source itself is uniform — a defective or poorly mixed source will masquerade as a detector problem. Novel and quality-assured flood sources continue to be studied precisely because source uniformity is part of the measurement.⁶

3. Acquire enough counts

Set the total count target from the per-pixel statistics, not by habit. If the matrix and count total leave only a few hundred counts per pixel, the flood is dominated by noise and the uniformity numbers are unreliable. Use high-count floods for uniformity-correction maps.¹³

4. Trend, don't just pass/fail

A single day's number can be within limits while a slow drift is underway. Record integral and differential uniformity over time and review the trend. The multi-site survey noted that trend analysis and periodic review were not common practice even where daily QC was — a gap that lets gradual degradation go unnoticed until it fails.⁴

5. Common pitfalls to avoid

• Reading only the integral value. A good integral number can hide a bad differential gradient that will ring in SPECT.¹
• Skipping the extrinsic flood. Collimator defects are invisible in an intrinsic test.
• Under-counting the flood. Too few counts per pixel make the metrics statistically meaningless.¹³
• Assuming a bad flood is the detector. Verify the source and the energy peak before condemning the camera.⁶
• Filing the number without trending. Drift is only visible across time.⁴

Regulatory Considerations

Gamma camera uniformity QC lives at the intersection of technical standards, accreditation requirements, and the radioactive-material license that authorizes the imaging program. The measurement methods are defined by NEMA NU 1-2018; the obligation to run a QC program comes from other authorities.¹

• Radioactive material license. Nuclear medicine imaging uses byproduct material regulated under 10 CFR Part 35 (medical use) and 10 CFR Part 20 (radiation protection), or the equivalent Agreement State program. While the NRC does not prescribe the exact uniformity limits, an effective instrument-QC program is part of operating a defensible medical-use program, and QC records support that posture during inspection.
• Accreditation. The ACR Nuclear Medicine Accreditation Program (and comparable programs) require documented uniformity and related QC as a condition of accreditation, which is frequently tied to reimbursement. The IAEA Human Health Series No. 6 provides detailed, internationally recognized procedures for acceptance, reference, and routine testing of scintillation cameras and SPECT systems that many programs use as a methodological basis.²
• Jurisdiction. Among the states DRPS serves, Florida, Maryland, Virginia, California, Nevada, Pennsylvania, New York, and New Jersey are NRC Agreement States that regulate medical use of byproduct material under their own radiation-control rules, while Washington DC and Delaware are regulated directly by the NRC. A facility should confirm which authority issues its license and which QC documentation expectations apply.

The practical result: NEMA defines how to measure uniformity, the manufacturer and accreditation program define what limits apply, and the medical physicist ties the two together into a documented program. For how this fits a broader compliance picture, see SPECT/CT quality control and our overview of dose calibrator quality control.

Frequently Asked Questions (FAQs)

What is gamma camera flood-field uniformity?

Flood-field uniformity describes how evenly a gamma camera counts a spatially uniform source of radiation across its detector. A perfectly uniform detector would record the same count density everywhere; real detectors show small variations from photomultiplier tube response, energy calibration, and collimator defects. Uniformity QC quantifies those variations so that non-uniformities are caught before they create image artifacts.

What is the difference between integral and differential uniformity?

Integral uniformity is the maximum deviation between the highest and lowest pixel counts across the field of view, expressed as a percentage. Differential uniformity measures the largest rate of change over any five contiguous pixels along rows and columns. Integral uniformity captures the worst overall variation; differential uniformity captures sharp local gradients that can mimic or hide lesions.

What is the difference between intrinsic and extrinsic uniformity testing?

Intrinsic uniformity is measured with the collimator removed, using a point source of Tc-99m at a distance, so it tests the detector and electronics alone. Extrinsic, or system, uniformity is measured with the collimator in place, usually with a Co-57 sheet source or a filled flood source, so it tests the whole imaging chain including the collimator. Both are used because collimator defects only appear in the extrinsic test.

How many counts are needed for a gamma camera flood?

Enough that statistical noise does not mask real non-uniformity. Because relative counting noise falls as one over the square root of the counts per pixel, achieving roughly one percent statistical fluctuation per pixel requires on the order of ten thousand counts in each pixel. Daily floods therefore typically accumulate several million to tens of millions of counts depending on matrix size, per NEMA and IAEA methodology.

Why is flood-field uniformity important for SPECT?

SPECT reconstructs many projection views into a tomographic image, and small uniformity errors are amplified into ring artifacts by the reconstruction process. Because of this sensitivity, SPECT systems use high-count uniformity correction floods and require tighter uniformity control than planar imaging alone, which is why uniformity is a core daily quality-control test.

How often should uniformity QC be performed?

Flood-field uniformity is typically checked daily before clinical imaging, with high-count uniformity correction floods acquired at the interval recommended by the manufacturer. Acceptance testing at installation establishes baseline NEMA values, and periodic physicist testing verifies ongoing performance against those baselines and against accreditation requirements.

Who sets gamma camera uniformity limits?

The measurement methods and definitions come from the NEMA NU 1 standard, while acceptance limits are set by the manufacturer's specifications, accreditation programs such as the ACR Nuclear Medicine Accreditation Program, and the facility's medical physicist. Daily pass/fail thresholds are usually vendor-based, with the physicist confirming that the system continues to meet its baseline and accreditation criteria.

Key Takeaways

• Uniformity is the most sensitive routine test. Many camera problems — PMT drift, energy peaking, linearity, collimator defects — appear first as a non-uniformity.³
• Read integral and differential uniformity together. A good integral value can hide a sharp local gradient that becomes a SPECT ring artifact.¹
• Intrinsic and extrinsic floods answer different questions. Only the extrinsic (collimator-in-place) test can reveal a collimator defect.
• Counting statistics govern validity. With relative noise of $1/\sqrt{N}$, meaningful floods need on the order of 10,000 counts per pixel, and high-count correction floods need far more.¹³
• SPECT is unforgiving. Reconstruction amplifies small uniformity errors, so SPECT demands tighter control and dedicated correction floods.¹⁴
• Trend the numbers. Slow drift is only visible over time, and trend review is a common gap in QC programs.⁴

Conclusion

Flood-field uniformity earns its place as the first test of the day because it is a broad, sensitive indicator of gamma camera health. Understanding the physics — that the reliability of the numbers comes from counting statistics, that integral and differential metrics capture different failure modes, and that SPECT reconstruction amplifies what planar imaging tolerates — turns a routine checkbox into a genuine early-warning system. A program built on NEMA definitions, adequate count statistics, both intrinsic and extrinsic floods, and disciplined trending catches problems while they are small, protects image quality, and stands up to accreditation and inspection.¹²³

How DRPS Can Help

Diagnostic Radiation Physics Services supports nuclear medicine facilities with gamma camera and SPECT/CT acceptance testing, uniformity and full NEMA performance evaluation, uniformity-correction flood review, QC-program design and trending, staff training, and accreditation preparation through our PET/CT and nuclear medicine physics, accreditation support, and medical physicist consulting services, all performed by board-certified medical physicists.

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

A strong uniformity program is not just about passing the daily check. It is about ensuring that every planar and SPECT study a facility produces is free of the artifacts that a drifting detector would otherwise create.

Related Resources

• Gamma camera NEMA NU-1 performance testing
• SPECT center of rotation QC
• SPECT/CT quality control
• PET/CT daily QC and calibration
• Dose calibrator quality control
• PET/CT and nuclear medicine physics
• Accreditation support

References

1. National Electrical Manufacturers Association. NEMA NU 1-2018: Performance Measurements of Gamma Cameras. Rosslyn, VA: NEMA; 2018. nema.org
2. International Atomic Energy Agency. Quality Assurance for SPECT Systems. IAEA Human Health Series No. 6. Vienna: IAEA; 2009. iaea.org
3. Zanzonico P. Routine quality control of clinical nuclear medicine instrumentation: a brief review. Journal of Nuclear Medicine. 2008;49(7):1114-1131. doi:10.2967/jnumed.107.050203. PubMed
4. Ramkishore N, Crocker J, Martin R, Yap KS, Brady Z. A survey of gamma camera and SPECT/CT quality control programs across a sample of public hospitals in Australia. Physical and Engineering Sciences in Medicine. 2024;47(3):1153-1166. doi:10.1007/s13246-024-01436-7. PubMed
5. Gear JI, Cummings C, Sullivan J, et al. Radioactive 3D printing for the production of molecular imaging phantoms. Physics in Medicine & Biology. 2020;65(17):175019. doi:10.1088/1361-6560/aba40e. PubMed
6. Saxena SK, Kumar Y, Malpani B, Rakshit S, Dash A. Development of a phantom and assessment of Ce-141 as a surrogate radionuclide for flood field uniformity testing of gamma cameras. Applied Radiation and Isotopes. 2016;112:115-121. doi:10.1016/j.apradiso.2016.03.014. PubMed
7. Pandey AK, Sharma PD, Kumar JP, et al. Calculating Gamma Camera Uniformity Parameters: Beyond the Vendor-specific Protocol. Indian Journal of Nuclear Medicine. 2017;32(4):279-282. doi:10.4103/ijnm.IJNM_67_17. PubMed
8. U.S. Nuclear Regulatory Commission. 10 CFR Part 35: Medical Use of Byproduct Material. ecfr.gov
9. American College of Radiology. Nuclear Medicine and PET Accreditation. acraccreditation.org