Quality Control
Snapshot
Reliable images depend on routine quality control of the gamma camera, PET scanner and dose calibrator, at defined frequencies. Daily checks (energy peak, uniformity/normalisation) and periodic checks (resolution, centre-of-rotation, sensitivity, phantom image quality) detect drift before it affects patients.
QC follows manufacturer and NEMA/EANM protocols. Failing QC (non-uniformity, COR error, dose-calibrator inaccuracy) directly produces artefacts or incorrect administered activity.
DailyUniformity / normalisation
NEMAPerformance standard
Dose calibratorActivity accuracy
Reference values
- NEMA NU 1 uniformity: IU = 100 × (max − min)/(max + min) over the UFOV or CFOV (75% of UFOV linear size), after 9-point smoothing on 6.4 mm pixels; DU uses the worst 5-pixel run.
- Flood counts (AAPM Report 177): NEMA 30 million (10 000 counts/pixel); routine intrinsic 10 million; routine extrinsic 5 million.
- Noise alone gives a baseline IU of 1.6% at 10 000 counts/pixel, 2.7% at 3 333 and 3.9% at 1 667.
- Routine action level: IU over the UFOV below 5% for 5-million-count floods.
- Intrinsic resolution 3–4 mm FWHM with ⁹⁹ᵐTc; nonlinearity under 1 mm; energy resolution 9–10% (refer for service if above 11%).
- PET (AAPM TG-126): uniform-phantom SUV 1.0 within 0.90–1.10; resolution and sensitivity within ±5% of baseline; PET/CT registration within 1 voxel.
- Dose calibrator (AAPM Report 181): constancy ±5%, accuracy ±5%, linearity ±5%, reproducibility ±1%.
Worked example
Energy resolution from FWHM
Given. ¹³⁷Cs 662 keV photopeak with FWHM = 55 keV.
- Energy resolution = FWHM/E × 100 = 55/662 × 100.
Answer. 8.3%; resolution in percent improves at higher photon energy.
Worked example
Integral and differential uniformity
Given. Smoothed 64×64 extrinsic flood (5 million counts). UFOV maximum 10 450, minimum 9 620 counts/pixel. Worst 5-pixel run in any row or column: 10 300 and 9 900.
- IU = 100 × (10 450 − 9 620)/(10 450 + 9 620) = 100 × 830/20 070 = 4.1%.
- DU = 100 × (10 300 − 9 900)/(10 300 + 9 900) = 2.0%.
- Noise alone contributes a baseline of about 3.9% at this count density.
Answer. IU 4.1% passes the 5% routine limit, but most of it is statistical: a genuine defect is better judged visually and on a higher-count flood.
Gamma camera / SPECT
- Daily: energy peak check and flood uniformity (extrinsic with the clinical collimator, or intrinsic).
- Weekly: spatial resolution and linearity with a quadrant bar phantom.
- Monthly (or as the manufacturer advises): centre of rotation and multi-head registration. Quarterly: SPECT phantom, with CT attenuation correction on SPECT/CT, and extrinsic floods of every collimator in use (AAPM Report 177).
PET
- Daily detector check to the manufacturer's protocol (⁶⁸Ge cylinder, ²²Na point source or intrinsic ¹⁷⁶Lu background); renormalise when it drifts beyond limits.
- Cross-calibration with the dose calibrator at least quarterly and after hardware or software changes; image quality and recovery coefficients annually (EANM).
- SUV accuracy also depends on synchronised clocks across calibrator, injection room and scanner.
Dose calibrator
- AAPM Report 181: background and constancy each day of use; accuracy, reproducibility and linearity at acceptance, after repair and annually; geometry (container and volume) at acceptance.
- The older US NRC schedule (quarterly linearity, annual accuracy, geometry at installation) is still followed in many departments; UK practice follows NPL Good Practice Guide 93.
- Traceable reference sources (⁵⁷Co, ¹³³Ba, ¹³⁷Cs) are used; see counting statistics and the dose calibrator.
Pitfalls
- Skipped uniformity QC leads to ring or field artefacts mistaken for pathology.
- COR error causes tomographic blurring or doughnut artefacts.
- Dose-calibrator drift causes systematic dosing and SUV error.
In the clinic — why the physics matters
- Resolution worsens with source-to-detector distance and in denser media — keep the organ close to the collimator.
- Uniformity correction prevents ring/bullseye artefacts in SPECT, which reconstruction amplifies.
- The monthly centre-of-rotation (COR) check keeps SPECT projections aligned — COR error → ring/doughnut artefacts.
- High count rates cause pulse pile-up → count loss and mis-positioned events, degrading dynamic/first-pass studies.
Routine QC schedule (summary)
| System | Test | Frequency | Reference / action level |
|---|---|---|---|
| Gamma camera | Peak and flood uniformity | Daily | IU <5% (5 M counts, UFOV) |
| Gamma camera | Bar phantom (resolution, linearity) | Weekly | Visual; nonlinearity <1 mm at acceptance |
| SPECT | COR and multi-head registration | Monthly or per manufacturer | Per manufacturer |
| SPECT/CT | Tomographic phantom with CT-AC | Quarterly | Compare with baseline |
| PET/CT | Daily detector QC | Daily | Per manufacturer |
| PET/CT | Cross-calibration / uniform SUV | Quarterly (EANM) | SUV 0.90–1.10 |
| Dose calibrator | Constancy | Each day of use | ±5% |
| Dose calibrator | Accuracy, linearity | Acceptance and annual | ±5% |
Common pitfalls & misconceptions
- Uniformity correction can't replace tuning — it handles minor PM-tube drift, not everything.
- The LSF FWHM underestimates degradation — it ignores scatter/septal-penetration tails; the MTF captures them.
- Intrinsic, collimator and system resolution aren't interchangeable — the collimator term usually dominates.
- Energy resolution isn't one fixed number — it improves with photon energy.
In depth
- NEMA NU 1 definitions: the UFOV is the area specified by the manufacturer, the CFOV its central 75% (linear); floods are reduced to about 6.4 mm pixels and smoothed with a 9-point (1-2-1) kernel before IU and DU are calculated.
- Intrinsic floods use a point source at least five UFOV diameters away (four is acceptable), shielded to reduce scatter, with background below 2% of the source rate (AAPM Report 177).
- Off-peak floods acquired with windows shifted to 126 and 154 keV reveal PMT decoupling and crystal hydration that a symmetric window can hide.
- New ⁵⁷Co sheet sources can contain ⁵⁶Co and ⁵⁸Co impurities that distort early extrinsic floods.
- System planar sensitivity for each collimator and nuclide should stay within 5% of baseline; a LEHR collimator gives roughly 75–90 cps/MBq.
- The quarterly SPECT phantom (e.g. Jaszczak, about 32 million counts) checks rods, spheres, uniformity and ring artefacts together, and on SPECT/CT also attenuation-correction accuracy.
- PET daily QC differs by vendor: ⁶⁸Ge cylinder, ²²Na point source or ¹⁷⁶Lu background. Trends should be reviewed at the annual physics survey.
- EANM routine QC recommendations (Busemann Sokole et al. 2010) and IAEA Human Health Series documents provide the European framework; AAPM Reports 126, 177 and 181 give US acceptance, annual and routine tests.
- US regulation (10 CFR 35.60) requires dose calibrators to be calibrated to nationally recognised standards or the manufacturer's instructions rather than prescribing a fixed schedule.
Sources: NEMA NU 1-2018 and NU 2-2018 · AAPM Reports 126 (2019), 177 (2019) and 181 (2012) · EANM routine QC 2010 (PMID 20130859) · EANM FDG guideline 2015 (PMID 25452219) · IAEA Human Health Series 1 and 6
Sources
- Busemann Sokole E, Płachcínska A, Britten A, et al. Routine quality control recommendations for nuclear medicine instrumentation. Eur J Nucl Med Mol Imaging. 2010;37:662-71.
- National Electrical Manufacturers Association. NEMA NU 1-2018: Performance measurements of gamma cameras. Rosslyn, VA: NEMA; 2018.
- National Electrical Manufacturers Association. NEMA NU 2-2018: Performance measurements of positron emission tomographs. Rosslyn, VA: NEMA; 2018.
- Halama JR, Graham D, Harkness BA, et al. AAPM Report No. 177: acceptance testing and annual physics survey recommendations for gamma camera, SPECT and SPECT/CT systems. College Park, MD: AAPM; 2019.
- Mawlawi OR, Kemp BJ, Jordan DW, et al. AAPM Report No. 126: PET/CT acceptance testing and quality assurance. College Park, MD: AAPM; 2019.
- Carey JE, Byrne P, DeWerd L, et al. AAPM Report No. 181: the selection, use, calibration, and quality assurance of radionuclide calibrators used in nuclear medicine. College Park, MD: AAPM; 2012.
- International Atomic Energy Agency. Quality Assurance for SPECT Systems. IAEA Human Health Series No. 6. Vienna: IAEA; 2009.
- International Atomic Energy Agency. Quality Assurance for PET and PET/CT Systems. IAEA Human Health Series No. 1. Vienna: IAEA; 2009.
- Boellaard R, Delgado-Bolton R, Oyen WJ, et al. FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0. Eur J Nucl Med Mol Imaging. 2015;42:328-54.