Gamma Camera Sensitivity QC

System sensitivity — the count rate a gamma camera records per unit of source activity — is a fundamental performance measure that ties directly to image quality, acquisition time, and quantitative accuracy. A camera with too little sensitivity forces longer scans or noisier images; one whose sensitivity has drifted quietly undermines quantitative SPECT.¹²

Measuring sensitivity correctly means decay-correcting the source activity, subtracting background, and comparing the result against the NEMA baseline and manufacturer specification. Trending it over time is one of the clearest early warnings of a degrading collimator, crystal, or detector.¹⁴ This guide covers what sensitivity is, how it is measured under NEMA NU 1, the collimator physics that sets it, a worked example, and how it fits into a defensible QC program.

Introduction

Every planar and SPECT image is built from counts, and sensitivity governs how quickly those counts accumulate. For a fixed activity and acquisition time, a more sensitive system delivers more counts, lower statistical noise, and either a shorter scan or a lower administered activity for the same image quality.¹³

Sensitivity is set almost entirely by the collimator, which rejects the overwhelming majority of emitted photons so that only those traveling nearly perpendicular to the detector are recorded. That rejection is what creates an image at all, but it also means a gamma camera detects only a tiny fraction of the photons a patient emits. Understanding and monitoring that fraction is central to nuclear medicine physics.³

This article defines sensitivity precisely, walks through the standardized NEMA measurement and a worked calculation, explains how collimator geometry determines the value, and shows how sensitivity trending protects both image quality and quantitative accuracy in a clinical program.

Topic Explanation

What is system sensitivity?

System sensitivity is the recorded count rate per unit of source activity, expressed in counts per second per megabecquerel (cps/MBq) or, in traditional units, counts per minute per microcurie (cpm/µCi). Formally, for a net count rate $R$ from a source of activity $A$:

The count rate must be background-subtracted, and the activity must be the value present at the time of acquisition, not at an earlier assay. Sensitivity characterizes the whole imaging chain — collimator, crystal, light guide, photomultiplier tubes, and pulse-height analyzer window — but for a properly peaked camera it is dominated by the collimator's geometric efficiency.¹³

NEMA NU 1 defines the standardized conditions for measuring system planar sensitivity so that results are comparable between systems and over time, using a source of known activity, a defined geometry, and a specified energy window.¹ The 2018 edition, NEMA NU 1-2018, is the version in force.¹

Sensitivity, resolution, and the collimator trade-off

Sensitivity never stands alone; it is locked in a trade-off with spatial resolution that the collimator design sets. Larger or shorter collimator holes admit more photons — raising sensitivity — but accept a wider range of incidence angles, blurring the image. Smaller or longer holes sharpen resolution at the cost of counts. No parallel-hole collimator escapes this inverse relationship.³

Medium-energy (MEGP) and high-energy (HEGP) collimators use thicker septa for higher-energy isotopes such as gallium-67, indium-111, or iodine-131; for technetium-99m specifically they are less efficient than the low-energy set and are chosen for septal penetration reasons, not to increase Tc-99m sensitivity.³

Why sensitivity is monitored

Sensitivity is monitored because a change in it signals a real, physical problem: a damaged or wrong collimator, a hydrated or cracked sodium iodide crystal, drifting photomultiplier gain, or an energy window that is off-peak. Because sensitivity is a single robust number, it trends cleanly, and a gradual decline is often visible before it degrades clinical images.²⁴

Key Technical Principles

Decay correction of the source

The source activity used in the sensitivity calculation must be corrected from the assay time to the acquisition time using the exponential decay law:

where $A_0$ is the assayed activity, $t$ is the elapsed time, and $T_{1/2}$ is the physical half-life — about 6.0 hours for technetium-99m and about 271.8 days for cobalt-57, the two radionuclides most often used for these measurements. Skipping or mis-timing this correction is one of the most common sources of error in a sensitivity measurement.¹

Collimator geometric efficiency

For a parallel-hole collimator, the fraction of emitted photons that reach the detector — the geometric efficiency — is well approximated by:

where $d$ is the hole diameter, $l_{\text{eff}}$ is the effective hole length (the physical length reduced for septal penetration), $t$ is the septal thickness, and $K$ is a shape constant of about 0.24–0.26 depending on hole packing. Two features of this expression explain clinical behavior: efficiency scales with the square of the hole diameter (bigger holes, far more counts) and it is essentially independent of source-to-collimator distance for a parallel-hole collimator, so planar count rate in air does not fall off with distance the way point-source intensity otherwise would.³

System sensitivity test conditions

A defensible sensitivity measurement standardizes the conditions so the result is reproducible and comparable.

Worked sensitivity example

Consider a system sensitivity measurement with technetium-99m:

• Assayed activity: $A_0=40.0$ MBq
• Elapsed time from assay to acquisition: $t=2.0$ h
• Acquisition time: 300 s
• Net (background-subtracted) counts: 1,050,000

First, decay-correct the activity to acquisition time using $T_{1/2}=6.0$ h:

Next, compute the net count rate:

Finally, the system sensitivity:

This value is then compared against the acceptance baseline and the manufacturer's specification for that collimator. If the current reading had instead come out well below baseline, the finding would prompt investigation of the collimator, crystal, peaking, and window before the system's quantitative calibration could be trusted.¹⁴

Clinical Impact

Sensitivity stability is the foundation of quantitative SPECT. Converting reconstructed counts into activity concentration requires a calibration factor — counts per second per unit activity — that is nothing other than a system sensitivity measurement. If sensitivity drifts, the calibration is wrong and every derived quantity drifts with it.⁵⁷

Published work bears this out. A study of SPECT/CT camera stability found that technetium-99m calibration factors and cobalt-57 daily-flood system sensitivity agreed to better than 3%, and that the camera was stable within 3% over an 18-month period — establishing daily flood-based sensitivity as a reliable tracker of quantitative stability.⁵ Clinical quantitative SPECT/CT using resolution recovery, attenuation, and scatter correction has been shown to recover absolute activity concentration to within a few percent when the calibration is sound.⁷ Cross-calibration studies further show that the choice of calibration source and geometry — cobalt-57 standard versus technetium-99m — affects the stability of the calibration factor, with the NEMA system planar sensitivity serving as the reference.⁸

The practical consequence: a nuclear medicine program that tracks sensitivity is protecting not only image quality and scan time but the validity of any quantitative result it reports.⁶⁸

Practical Optimization Tips

Peak before you measure

A sensitivity measurement inherits any peaking error. Confirm the photopeak and energy window are correct for the radionuclide — 140 keV for technetium-99m, 122 keV for cobalt-57 — before acquiring, so an off-peak window is not misread as low sensitivity.⁴

Get the activity right

The measurement is only as good as the assayed activity. Use a dose calibrator with current constancy, accuracy, and linearity checks, record the assay time precisely, and decay-correct to the acquisition time. A ten-minute timing error on a technetium-99m source is a real percentage error in the result.¹

Subtract background and match geometry

Acquire and subtract a background frame, and reproduce the acceptance-test geometry — source-to-detector distance, source preparation, and window — so the comparison against baseline is meaningful. Confirm the source itself is what you think it is; a spilled or partially assayed source corrupts the number.¹

Trend, don't just pass

Record the acceptance-testing sensitivity and manufacturer specification as references, and trend each subsequent measurement. A single in-specification value is reassuring, but a slow decline across measurements is the signal that catches a developing collimator or crystal problem early — and trend review is a common gap in QC programs.²⁴

Regulatory Considerations

Gamma camera sensitivity sits at the intersection of technical performance standards, accreditation, and the radioactive-material license. The measurement method comes from NEMA; the expectation to perform it comes from accreditation and the medical physicist's survey; the broader program lives under the medical-use license.

• Technical standard. NEMA NU 1-2018 defines how system sensitivity is measured, so values are comparable across vendors and over time.¹ International guidance in IAEA Human Health Series No. 6 and IAEA-TECDOC-602 describes acceptance and routine QC for scintillation cameras and SPECT systems, including sensitivity.²⁹
• Acceptance and survey. AAPM Report No. 177 (Task Group 177) sets out acceptance-testing and annual performance-evaluation recommendations for gamma camera, SPECT, and SPECT/CT systems, including sensitivity, performed by a qualified medical physicist.⁴
• License context. The camera is a device, not radioactive material, so sensitivity itself is a NEMA/manufacturer and physicist-survey test rather than a specific NRC-mandated measurement. The medical-use license under 10 CFR Part 35 (or the equivalent Agreement State program) governs the surrounding radiation-safety and instrument-QA program in which the camera operates.¹⁰ In Florida, radioactive-material medical use is administered by the state under Florida Administrative Code Chapter 64E-5; DRPS also serves Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware. Always confirm requirements with the authority having jurisdiction.

For related instrument QC, see our guides to gamma camera uniformity QC and dose calibrator quality control.

Frequently Asked Questions (FAQs)

What units is sensitivity reported in?

Most commonly counts per second per megabecquerel (cps/MBq), or counts per minute per microcurie (cpm/µCi) in traditional units. What matters is consistency: the same units, geometry, and radionuclide must be used when comparing against baseline.¹

Is planar sensitivity the same as SPECT sensitivity?

They are related but not identical. Planar system sensitivity is the standardized NEMA measurement; SPECT (tomographic) sensitivity and the quantitative calibration factor derive from it but also depend on acquisition and reconstruction settings. Programs track the planar value as the stable reference.¹⁸

Why does distance not change parallel-hole sensitivity much?

For a parallel-hole collimator, as a source moves away the count rate per detector area falls, but the area of the detector that can "see" the source grows in proportion, so the total recorded count rate stays nearly constant. This is a direct consequence of the geometric-efficiency expression.³

What causes a sudden drop in sensitivity?

Common causes include the wrong collimator mounted, a damaged or hydrated crystal, off-peak energy windows, photomultiplier gain drift, or an error in the assayed activity or its decay correction. Ruling out the measurement error first is essential before blaming the detector.⁴

Who should perform sensitivity testing?

A qualified or board-certified medical physicist performs sensitivity as part of acceptance testing and the periodic performance evaluation, using calibrated instrumentation and standardized conditions.⁴

Key Takeaways

• Sensitivity is the net count rate per unit source activity, S = R / A, in cps/MBq.¹
• The source activity must be decay-corrected to the acquisition time before dividing.¹
• Collimator geometric efficiency dominates sensitivity and trades off inversely with spatial resolution.³
• Parallel-hole sensitivity is essentially independent of source distance in air.³
• NEMA NU 1-2018 standardizes the measurement; AAPM TG-177 sets acceptance and survey expectations.¹⁴
• Sensitivity stability underpins quantitative SPECT — calibration factors have been shown stable within about 3% when tracked.⁵
• Trend sensitivity against baseline; a slow decline is an early warning of collimator or crystal problems.²⁴

How DRPS Can Help

Diagnostic Radiation Physics Services (DRPS) supports nuclear medicine facilities across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware with PET/CT and nuclear medicine physics, gamma camera and SPECT/CT acceptance testing, full NEMA performance evaluation including sensitivity, quantitative SPECT calibration, QC program design and trending, and accreditation preparation — all performed by board-certified medical physicists.

A strong sensitivity program is not just about passing a specification. It is about confirming that the camera is converting activity into counts as expected, that the quantitative calibration can be trusted, and that a developing detector or collimator problem is caught while it is still small.

Conclusion

System sensitivity is a compact, physically meaningful measure of how efficiently a gamma camera turns activity into counts. Measured correctly — with a decay-corrected known activity, subtracted background, and standardized NEMA conditions — it supports faster scans, lower administered activity, and, critically, trustworthy quantitative SPECT. Trended against an acceptance baseline, sensitivity becomes one of the most reliable early-warning signals in a nuclear medicine QC program.¹⁴⁵

Related Resources

• Gamma camera uniformity QC
• Gamma camera energy resolution QC
• Gamma camera NEMA NU-1 performance testing
• Gamma camera collimator selection
• Quantitative SPECT/CT calibration
• PET/CT and nuclear medicine physics services

References

1. National Electrical Manufacturers Association. Performance Measurements of Gamma Cameras. NEMA Standards Publication NU 1-2018. 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. International Atomic Energy Agency. Nuclear Medicine Physics: A Handbook for Teachers and Students. Vienna: IAEA; 2014. iaea.org
4. Halama JR, Madsen MT, et al. Acceptance Testing and Annual Physics Survey Recommendations for Gamma Camera, SPECT, and SPECT/CT Systems. AAPM Report No. 177 (Task Group 177). College Park, MD: AAPM; 2019. aapm.org
5. McDougald WA, Miyaoka RS, Alessio AM, Harrison RL, Lewellen TK. A study of SPECT/CT camera stability for quantitative imaging. EJNMMI Physics. 2016;3(1):14. doi:10.1186/s40658-016-0150-7. doi.org
6. Strugari ME, DeBay DR, Beyea SD, Brewer KD. NEMA NU 1-2018 performance characterization and Monte Carlo model validation of the Cubresa Spark SiPM-based preclinical SPECT scanner. EJNMMI Physics. 2023;10(1):35. doi:10.1186/s40658-023-00555-6. doi.org
7. Zeintl J, Vija AH, Yahil A, Hornegger J, Kuwert T. Quantitative accuracy of clinical 99mTc SPECT/CT using ordered-subset expectation maximization with 3-dimensional resolution recovery, attenuation, and scatter correction. Journal of Nuclear Medicine. 2010;51(6):921-928. doi:10.2967/jnumed.109.071571. doi.org
8. Miyaji N, Miwa K, Motegi K, et al. Validation of cross-calibration schemes for quantitative bone SPECT/CT using different sources under various geometric conditions. Nihon Hoshasen Gijutsu Gakkai Zasshi. 2017;73(6):443-450. doi:10.6009/jjrt.2017_JSRT_73.6.443. doi.org
9. International Atomic Energy Agency. Quality Control of Nuclear Medicine Instruments. IAEA-TECDOC-602. Vienna: IAEA; 1991. iaea.org
10. U.S. Nuclear Regulatory Commission. 10 CFR Part 35, Medical Use of Byproduct Material. nrc.gov