Is attenuation correction required for accreditation?

Attenuation correction is endorsed by professional society position statements as improving specificity when performed with proper quality control, and nuclear cardiology laboratory accreditation evaluates image quality, processing, and interpretation. The medical physicist verifies that the transmission or CT subsystem, the attenuation map, and registration are functioning correctly as part of the quality-control program.

Cardiac SPECT Attenuation Correction & Artifacts

Q: How does attenuation correction work in cardiac SPECT?

Attenuation correction builds a map of how strongly each location in the chest attenuates photons, then mathematically compensates the emission data for that loss. The map comes either from a transmission scan using an external radionuclide source or, in modern SPECT/CT systems, from a low-dose CT scan converted to attenuation coefficients at the relevant photon energy.

Q: What is the difference between line-source and CT-based attenuation correction?

Older systems used an external sealed radionuclide transmission source (for example a scanning line source) to measure attenuation directly at a gamma energy. Modern SPECT/CT uses a low-dose CT scan, which is faster and higher quality but must be scaled from CT x-ray energies to the gamma energy and must be accurately registered to the emission scan.

Q: How do you tell an attenuation artifact from real scar?

Gated SPECT is the key tool. An attenuation artifact is a fixed count reduction with preserved regional wall motion and wall thickening, whereas infarct scar shows a fixed defect with reduced or absent wall motion. Attenuation correction, prone imaging, and the characteristic location and shape of the defect also help distinguish artifact from disease.

Q: Does attenuation correction ever create its own artifacts?

Yes. The most common problem is misregistration between the CT attenuation map and the emission data, often from respiratory or patient motion, which can create or mask apparent defects. CT truncation, beam hardening, and metal can also distort the map. Quality control of the registration is essential before trusting a corrected study.

Q: Does prone imaging replace attenuation correction?

No; they are complementary. Prone positioning shifts the diaphragm and heart to reduce inferior-wall attenuation and motion, and is a low-cost option for cameras without a transmission or CT source. Attenuation correction with a measured map addresses attenuation more completely and in all walls, so many laboratories use gating plus either prone imaging or attenuation correction.

Soft-tissue attenuation is the single most common cause of false-positive myocardial perfusion SPECT: photons from the inferior wall are absorbed by the diaphragm and those from the anterior wall by breast tissue, producing fixed "defects" where blood flow is actually normal. Attenuation correction — using a transmission or CT-derived map of how the chest absorbs photons — plus gated wall-motion analysis is how nuclear cardiology separates these artifacts from real coronary disease.¹²³

A perfusion defect that is really a shadow of the patient's own anatomy is a diagnostic trap: it can send a healthy patient to catheterization or mask the very disease the scan was ordered to find. This article explains the attenuation physics that creates the artifact, the two families of attenuation correction (radionuclide transmission and CT-based), a worked calculation of how much signal the inferior wall loses, how gated SPECT and prone imaging help, and the quality control that keeps attenuation correction from introducing new errors. DRPS supports nuclear cardiology programs through PET/CT and nuclear medicine physics and medical physicist consulting across Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware.

Introduction

Myocardial perfusion imaging (MPI) measures the relative distribution of a radiotracer in the heart muscle, and it assumes that a uniformly perfused myocardium produces uniform counts. Attenuation breaks that assumption. A photon emitted from the heart must traverse several centimeters of tissue — lung, muscle, bone, breast, and diaphragm — before reaching the detector, and some fraction is absorbed or scattered away. Because the overlying tissue is not uniform around the heart, the count loss is not uniform either, and the result can look exactly like a regional perfusion defect.¹²

The clinical stakes are high. Attenuation artifacts are a principal reason that stress MPI historically had good sensitivity but limited specificity, and they are the main target of attenuation correction, gating, and positioning strategies.²⁶ This guide works through the physics, the correction methods, artifact recognition, practical optimization, and the regulatory and quality-control context that a medical physicist and nuclear cardiology laboratory must manage.

Topic Explanation

What is attenuation in cardiac SPECT?

Attenuation is the loss of gamma photons as they pass through tissue, by photoelectric absorption and Compton scatter, before they can be detected. For the technetium-99m (Tc-99m) agents that dominate modern MPI, the gamma energy is 140 keV; for thallium-201 (Tl-201), the useful x-rays are near 70–80 keV, where tissue attenuation is even stronger.¹⁷ The deeper a photon's origin and the more tissue it must cross, the fewer photons survive to form the image.

Key terms used throughout this guide:

Linear attenuation coefficient ( $μ$ ) — the probability per unit path length that a photon is removed from the beam; it depends on photon energy and tissue type.
Transmission scan — a measurement of how strongly the body attenuates photons, used to build an attenuation map.
Attenuation map ( $μ$ -map) — a spatial map of attenuation coefficients through the chest, used to correct the emission data.
Gated SPECT — acquisition synchronized to the ECG so regional wall motion and thickening can be assessed alongside perfusion.

Where attenuation artifacts appear

The two classic patterns follow anatomy:¹²

Inferior-wall attenuation from the left hemidiaphragm and abdominal contents, most pronounced in larger men and with a high diaphragm; it mimics an inferior or inferobasal defect.
Anterior-wall attenuation from breast tissue, most pronounced in women and dependent on breast position, which can shift between stress and rest and move the apparent defect.

Lateral-wall attenuation (obesity, arm position) and general count loss from large body habitus round out the picture. Because breast position is variable, an anterior defect that changes location or depth between acquisitions is a strong clue that attenuation, not disease, is responsible.²

Key Technical Principles

The attenuation law

Photon transmission through a uniform thickness $x$ of tissue with linear attenuation coefficient $μ$ follows the exponential attenuation law:¹⁷

I = I_{0} e^{- μx}

At 140 keV, the linear attenuation coefficient of soft tissue is approximately $μ \approx 0.15 cm^{- 1}$ , which corresponds to a half-value layer of:

HVL = \frac{ln 2}{μ} = \frac{0.693}{0.15} \approx 4.6 cm

So roughly every 4.6 cm of soft tissue halves the number of 140 keV photons reaching the detector.⁷

Worked example: why the inferior wall looks "cold"

Consider two myocardial regions emitting equal activity but seen through different tissue depths. Suppose anterior-wall photons traverse about 5 cm of tissue to the anterior detector position, while inferior-wall photons traverse about 10 cm because of the diaphragm and abdomen. The surviving fractions are:

\frac{I _{ant}}{I _{0}} = e^{- 0.15 \times 5} = e^{- 0.75} \approx 0.47

\frac{I _{inf}}{I _{0}} = e^{- 0.15 \times 10} = e^{- 1.5} \approx 0.22

The ratio of detected counts is:

\frac{I _{inf}}{I _{ant}} = \frac{0.22}{0.47} \approx 0.47

The inferior wall delivers less than half the counts of the anterior wall despite identical perfusion. On a relative-scaled perfusion map, that reads as a marked inferior "defect" — a pure artifact of depth-dependent attenuation.¹² This is the quantitative core of why attenuation correction matters.

Transmission-based attenuation correction

Transmission-based correction measures attenuation directly using an external sealed radionuclide source — historically a scanning line source, a fixed line source, or a sheet source — positioned opposite the detector. The source projects photons through the patient, and the measured transmission builds a $μ$ -map at (or near) the gamma energy. Gadolinium-153, with emissions near 100 keV, was a common transmission source because its energy is close enough to 140 keV to require only modest scaling.¹² Transmission correction measures attenuation at gamma energies but is limited by transmission-source decay, lower-resolution maps, and cross-contamination between emission and transmission windows.

CT-based attenuation correction (SPECT/CT)

Modern hybrid SPECT/CT systems derive the attenuation map from a low-dose CT scan. The CT provides a high-quality, low-noise map of x-ray attenuation, which is then converted (via a bilinear scaling) from the effective CT energy to the gamma energy of the tracer to produce the $μ$ -map used for correction.¹⁵ CT-based correction is fast and anatomically detailed, and the added CT dose is modest — on the order of about 0.5 mSv of effective dose for the CT component of a cardiac SPECT/CT, small relative to the emission dose.⁵

The central requirement of CT-based correction is registration: the CT $μ$ -map and the emission data must align. Because the CT is acquired over seconds and the SPECT over minutes, respiratory position and patient motion can misalign them, and a misregistered map is the most common source of attenuation-correction artifacts.¹²

Chang's method and the role of a measured map

For a uniform attenuator, a first-order analytic correction (Chang's method) multiplies each reconstructed pixel by a factor derived from the average attenuation path over all projection angles:⁴

C (x, y) = [\frac{1}{N} i = 1 \sum N e^{- μ l_{i} (x, y)}]^{- 1}

where $l_{i}$ is the path length from pixel $(x, y)$ to the body edge at projection angle $i$ . Chang's uniform correction is reasonable for the brain or abdomen but not for the thorax, where lung, soft tissue, and bone make attenuation strongly nonuniform. That nonuniformity is exactly why cardiac SPECT requires a measured map — from transmission or CT — rather than a single assumed $μ$ .¹⁴

Comparison of attenuation-management strategies

Strategy	What it does	Strengths	Limitations / QC watch points
Radionuclide transmission (line/sheet source)	Measures a gamma-energy $μ$ -map directly	Energy close to emission; no CT dose	Source decay, noisy/low-res map, window crosstalk
CT-based (SPECT/CT)	Low-dose CT scaled to gamma energy	Fast, detailed, low noise, adds about 0.5 mSv	Emission–CT misregistration; truncation; beam hardening; metal
Prone imaging	Repositions diaphragm and heart	Low cost; reduces inferior attenuation and motion	Anterior attenuation may change; limited coverage
Gated SPECT (adjunct)	Adds wall motion and thickening	Separates artifact from scar	Needs adequate counts and a stable rhythm

No single row is a complete answer; contemporary practice pairs gating with either a measured attenuation map or prone imaging.¹²³

Clinical Impact

Attenuation correction and gating together raise the specificity and the normalcy rate of MPI, reducing unnecessary downstream catheterization. Professional-society position statements conclude that attenuation correction, performed with appropriate quality control, improves the interpretive certainty and specificity of SPECT MPI and is a recommended option for laboratories that can implement it correctly.²³ The practical payoff is fewer equivocal reads and fewer false-positive inferior and anterior defects in exactly the patients — larger men, women with dense breast tissue — in whom attenuation is worst.²

The flip side is that attenuation correction is not a cure-all and can mislead if applied without quality control. A misregistered CT map can introduce an apparent defect or erase a true one, so the corrected and non-corrected images are always read together, with the raw rotating projection cinema reviewed for motion and the fused CT checked for alignment.¹² Gated wall motion remains the decisive adjunct: a fixed perfusion reduction with normal wall motion and thickening is the signature of attenuation, while a fixed defect with a wall-motion abnormality points to infarction.¹ For the gating side of this workflow, see our guide to gated SPECT MPI and LVEF quantification.

Practical Optimization Tips

Acquire so artifacts are recognizable

Always gate when count statistics allow; wall motion and thickening are the primary way to separate attenuation from scar.¹
Review the rotating projections (the raw cine) before processing to catch patient motion, breast position, and "upward creep" of the heart after exercise.²
Use prone or upright/supine combinations to shift attenuating structures when a transmission or CT source is unavailable; a defect that normalizes prone is characteristic of diaphragmatic attenuation.¹²

Get attenuation correction right

Verify emission–CT registration on every SPECT/CT study, in all three planes, before accepting the corrected images; realign or repeat if the heart and CT border do not coincide.¹²
Read corrected and non-corrected images together, with gated data, rather than trusting either alone; discordance is a prompt to investigate, not to pick a favorite.²³
Watch for CT-specific pitfalls — truncation of large patients, beam hardening, and metal (pacemakers, leads) — that distort the $μ$ -map.¹
Maintain the transmission source where used: track its decay and replace on schedule so the map is not progressively degraded.¹

Build it into the physics QC program

A qualified medical physicist should include the attenuation-correction subsystem in acceptance and routine quality control: transmission-source uniformity and decay, CT calibration and registration accuracy, and the behavior of the correction on a uniform phantom. This sits alongside routine camera QC such as uniformity, center of rotation, and energy calibration, and connects to quantitative work described in quantitative SPECT/CT calibration.

Regulatory Considerations

Cardiac SPECT and SPECT/CT sit at the intersection of NRC (or Agreement State) materials regulation for the radiopharmaceutical and state radiation-machine rules for the CT subsystem, with professional-society guidelines governing clinical quality.

Byproduct material (the tracer). Possession and medical use of Tc-99m and Tl-201 agents fall under 10 CFR Part 35 (or the equivalent Agreement State program), with dose limits under 10 CFR Part 20. The authorized user, written directive where applicable, and radiation safety program apply as for any nuclear medicine procedure.
CT subsystem (SPECT/CT). The CT used for attenuation correction and localization is a radiation-producing machine regulated by the state radiation-control program; it requires acceptance testing and a medical physicist's evaluation like any clinical CT, even at the low doses used for attenuation maps.⁵
Professional guidelines and accreditation. The ASNC/SNMMI SPECT MPI guidelines and the ASNC/SNM attenuation-correction position statements define acquisition, processing, correction, and interpretation expectations, and laboratory accreditation evaluates image quality and quality control.¹²³ The IAEA has published implementation guidance for SPECT MPI that reflects the same principles internationally.

Of the jurisdictions DRPS serves, Florida, Maryland, Virginia, California, Nevada, Pennsylvania, New York, and New Jersey are NRC Agreement States that license medical use under their own radiation-control rules, while Washington, DC and Delaware are regulated directly by the NRC for radioactive material. The CT side is regulated by each state's radiation-machine program. Always confirm requirements with the authority having jurisdiction. For the jurisdictional map, see our guide to NRC, Agreement States, and FDA jurisdiction.

Frequently Asked Questions (FAQs)

Why does soft-tissue attenuation cause false perfusion defects?

Gamma photons emitted deep in the body are absorbed and scattered before reaching the detector, so regions overlaid by more tissue produce fewer counts. The diaphragm attenuates the inferior wall and breast tissue attenuates the anterior wall, creating low-count regions that look like fixed defects even when perfusion is normal.¹²

How does attenuation correction work in cardiac SPECT?

It builds a map of how strongly each location in the chest attenuates photons, then compensates the emission data for that loss during reconstruction. The map comes from an external radionuclide transmission scan or, in SPECT/CT, from a low-dose CT scaled to the gamma energy.¹⁵

What is the difference between line-source and CT-based attenuation correction?

A radionuclide transmission source measures attenuation directly at a gamma energy but gives a noisier, lower-resolution map and decays over time. CT-based correction is faster and more detailed but must be scaled from x-ray to gamma energy and accurately registered to the emission scan.¹⁵

How do you tell an attenuation artifact from real scar?

Gated SPECT is decisive: an attenuation artifact is a fixed count reduction with preserved wall motion and thickening, whereas scar shows a fixed defect with reduced or absent wall motion. Defect location, shape, and response to attenuation correction or prone imaging also help.¹

Does attenuation correction ever create its own artifacts?

Yes — most often from misregistration between the CT map and the emission data due to motion or respiration, which can create or mask defects. Truncation, beam hardening, and metal can also distort the map, so registration quality control is essential before trusting a corrected study.¹²

Does prone imaging replace attenuation correction?

No; they are complementary. Prone positioning reduces inferior-wall attenuation and motion and is a low-cost option without a transmission or CT source, while a measured attenuation map addresses attenuation more completely in all walls. Many laboratories pair gating with prone imaging or attenuation correction.¹²

Key Takeaways

Attenuation follows $I = I_{0} e^{- μx}$ ; at 140 keV soft tissue has $μ \approx 0.15 cm^{- 1}$ and a half-value layer near 4.6 cm.⁷
Depth-dependent count loss makes the inferior wall deliver roughly half the counts of the anterior wall at equal perfusion — a pure artifact.¹
The thorax is nonuniform, so cardiac SPECT needs a measured $μ$ -map (transmission or CT), not Chang's uniform assumption.¹⁴
CT-based correction is fast and detailed but lives or dies on emission–CT registration; verify it on every study.¹²
Gated SPECT separates attenuation artifact (normal wall motion) from scar (abnormal wall motion); read corrected and non-corrected images together.¹³
Attenuation correction and gating improve specificity and the normalcy rate when backed by proper quality control.²³

Conclusion

Attenuation is not a flaw in the scanner; it is physics — the exponential loss of photons through tissue that the body imposes on every emitted gamma ray. In the chest, where attenuation is strongly nonuniform, that physics manufactures inferior and anterior "defects" that have nothing to do with coronary disease. The answer is not a single trick but a disciplined combination: a measured attenuation map from a transmission source or a registered CT, gated wall-motion analysis to separate artifact from scar, prone or combined positioning where helpful, and a quality-control program that verifies the correction is helping rather than harming. Applied this way, attenuation correction turns one of nuclear cardiology's oldest weaknesses into a measurable, correctable, and defensible part of the study.¹²³

How DRPS Can Help

Diagnostic Radiation Physics Services supports nuclear cardiology and SPECT/CT programs with PET/CT and nuclear medicine physics, gamma camera and SPECT/CT acceptance testing and quality control, attenuation-correction and registration verification, CT dose and image-quality evaluation, accreditation support, and medical physicist consulting performed by board-certified medical physicists.

DRPS serves facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware. The goal is simple: perfusion reads that reflect blood flow, not body habitus.

Related Resources

References

Dorbala S, Ananthasubramaniam K, Armstrong IS, et al. Single photon emission computed tomography (SPECT) myocardial perfusion imaging guidelines: instrumentation, acquisition, processing, and interpretation. J Nucl Cardiol. 2018;25(5):1784-1846. doi:10.1007/s12350-018-1283-y. PubMed
Hendel RC, Corbett JR, Cullom SJ, DePuey EG, Garcia EV, Bateman TM. The value and practice of attenuation correction for myocardial perfusion SPECT imaging: a joint position statement from the ASNC and the Society of Nuclear Medicine. J Nucl Cardiol. 2002;9(1):135-143. doi:10.1067/mnc.2002.120680. PubMed
Heller GV, Links J, Bateman TM, et al. American Society of Nuclear Cardiology and Society of Nuclear Medicine joint position statement: attenuation correction of myocardial perfusion SPECT scintigraphy. J Nucl Cardiol. 2004;11(2):229-230. doi:10.1016/j.nuclcard.2003.12.001. PubMed
Chang LT. A method for attenuation correction in radionuclide computed tomography. IEEE Trans Nucl Sci. 1978;25(1):638-643. doi:10.1109/TNS.1978.4329385. doi.org
Abdollahi H, Shiri I, Salimi Y, et al. Radiation dose in cardiac SPECT/CT: an estimation of SSDE and effective dose. Eur J Radiol. 2016;85(12):2257-2261. doi:10.1016/j.ejrad.2016.10.021. PubMed
Badheka AO, Hendel RC. Radionuclide cardiac stress testing. Curr Opin Cardiol. 2011;26(5):370-378. doi:10.1097/HCO.0b013e328349803b. PubMed
Cherry SR, Sorenson JA, Phelps ME. Physics in Nuclear Medicine. 4th ed. Philadelphia, PA: Elsevier Saunders; 2012. elsevier.com
Bushberg JT, Seibert JA, Leidholdt EM, Boone JM. The Essential Physics of Medical Imaging. 4th ed. Philadelphia, PA: Wolters Kluwer; 2021. wolterskluwer.com
National Institute of Standards and Technology. XCOM: Photon Cross Sections Database (mass attenuation coefficients for tissue). Gaithersburg, MD: NIST. nist.gov
International Atomic Energy Agency. Nuclear Cardiology: Guidance on the Implementation of SPECT Myocardial Perfusion Imaging. IAEA Human Health Series No. 23. Vienna: IAEA; 2016. iaea.org