Alicia sees a coloured myocardial perfusion image and assumes the scanner shines radiation through the heart like CT. Tricia notices the opposite geometry: the patient receives a tiny amount of radiotracer, and the detector waits for radiation emitted from inside the body. Kai Kai asks how those detected photons become a map of blood delivery.
Nuclear cardiac perfusion imaging works by administering a radiotracer whose delivery or uptake reflects myocardial blood flow, detecting radiation emitted as that tracer decays and reconstructing the detected events into images or quantitative flow measurements. SPECT detects gamma photons emitted directly by radionuclides. PET detects pairs of 511-keV photons produced when emitted positrons annihilate with electrons. Stress and rest acquisitions can compare perfusion states, while modern PET—and increasingly specialised SPECT—can estimate absolute myocardial blood flow and myocardial flow reserve.
This article complements How Echocardiography Works, How Cardiac MRI Works, How Cardiac CT Works and How Coronary Autoregulation Works. Here the reader job is tracer physics and perfusion measurement, not diagnosis or personal scan interpretation.
1. Nuclear imaging begins with a tracer inside the body
A radiotracer combines a radioactive isotope with a chemical form whose distribution follows a biological process.
For myocardial perfusion imaging, useful tracers reach heart muscle through coronary blood flow and either become temporarily retained or provide a time-varying signal related to delivery.
The image therefore maps the history of tracer transport rather than photographing coronary arteries directly.
2. Radioactive decay provides the detectable signal
Unstable radionuclides transform toward more stable nuclear states and release energy.
Different isotopes emit different kinds of radiation and have different physical half-lives.
Tracer choice therefore affects detector physics, timing, image quality, workflow and how faithfully uptake tracks myocardial blood flow.
3. SPECT detects gamma photons
Single-photon emission computed tomography uses radionuclides that emit gamma photons detectable outside the body.
Common myocardial perfusion SPECT agents include technetium-99m-labelled compounds such as sestamibi and tetrofosmin.
The 2025 review Radionuclide Tracers for Myocardial Perfusion Imaging and Blood Flow Quantification summarises the differing extraction and imaging properties of contemporary SPECT and PET tracers.
4. SPECT needs a collimator to know photon direction
A gamma camera can detect where a photon hits the detector but not automatically where it came from.
A physical collimator permits photons travelling along selected directions to reach the detector while absorbing many others.
This provides directional information but sacrifices many emitted photons, creating an important sensitivity–resolution trade-off.
5. Rotating SPECT projections become tomographic slices
Gamma-camera detectors acquire many two-dimensional projections around the chest.
Reconstruction algorithms combine those projections to estimate the three-dimensional distribution of tracer activity.
The resulting perfusion slices are therefore reconstructed activity maps rather than direct photographs of tissue colour.
6. PET detects annihilation photons in coincidence
Positron-emitting radionuclides release a positron. After travelling a short distance, the positron encounters an electron and the pair annihilates.
The annihilation produces two photons travelling in nearly opposite directions, each with energy near 511 keV.
When detectors record both photons within a very short timing window, the event is assigned to a line of response between those detectors.
7. PET uses electronic rather than physical collimation
Coincidence detection gives PET directional information without a lead collimator blocking most photons.
This contributes to PET’s high sensitivity and supports dynamic imaging for quantitative myocardial blood flow.
Resolution is still limited by detector size, positron range, non-collinearity and reconstruction physics.
8. Common PET perfusion tracers have different kinetics
Rubidium-82, nitrogen-13 ammonia and oxygen-15 water have all been used for myocardial perfusion PET, while fluorine-18 flurpiridaz is an important newer tracer.
Their half-lives, extraction fractions and retention mechanisms differ.
A 2026 review 18F-Flurpiridaz PET in myocardial perfusion imaging discusses how newer tracer physics and logistics differ from traditional PET agents.
9. Tracer uptake is not identical to blood flow
A tracer must be delivered by blood, extracted from blood and retained or measured according to its chemical behaviour.
At very high flows, some tracers do not increase tissue uptake in perfect linear proportion to flow because extraction becomes less complete.
Quantitative models therefore need tracer-specific extraction and kinetic assumptions rather than treating detector counts as litres per minute directly.
10. Relative perfusion compares regions within the same heart
Traditional perfusion imaging often scales myocardial activity to the region with the highest uptake.
Regions receiving less tracer then appear relatively reduced.
This is useful for spatial comparison but can hide globally reduced perfusion if all territories are affected similarly.
11. Absolute myocardial blood flow adds a physical rate
Dynamic PET acquires rapid frames as tracer arrives in blood and myocardium.
Time-activity curves from the arterial blood pool and myocardial tissue are fitted with tracer-kinetic models to estimate myocardial blood flow, commonly expressed in mL/min/g.
The 2025 review Quantification of myocardial blood flow using PET describes the contemporary role of absolute flow and flow reserve.
12. Flow reserve compares stress with rest
Myocardial flow reserve is typically calculated as hyperaemic or stress myocardial blood flow divided by resting myocardial blood flow.
The ratio asks how much the coronary circulation can increase flow above its baseline state.
It therefore relates directly to the physiological reserve concept developed in the Coronary Autoregulation and Cardiac Reserve pillars.
13. Stress creates a second operating state
Perfusion imaging can compare rest with exercise or pharmacologically induced hyperaemia.
The 2026 ASNC guideline for stress testing in SPECT and PET myocardial perfusion imaging reflects current practice standards for how those stress states are created and monitored clinically.
This educational article explains the measurement mechanism only and does not give patient preparation or stress-test instructions.
14. Attenuation changes detected counts
Photons emitted inside the chest can be absorbed or scattered before reaching the detector.
Different paths through lungs, soft tissue and bone have different attenuation.
Without correction, identical tracer concentrations in two regions can produce different detected counts simply because the photons travelled through different amounts of tissue.
15. CT attenuation maps help correct nuclear images
Hybrid SPECT/CT and PET/CT systems can acquire an x-ray CT attenuation map of the chest.
The reconstruction uses that map to estimate how likely photons are to be absorbed along different paths.
Registration matters: if the heart moves between CT and emission acquisition, the correction map can be misaligned with the tracer image.
16. Scatter creates misplaced or mis-energised events
A gamma photon can scatter inside the body and change direction before reaching a detector.
If accepted as though it travelled straight, it contributes activity to the wrong spatial estimate.
Energy windows and model-based corrections reduce but do not completely eliminate scatter effects.
17. PET has random and scattered coincidences
Not every pair of nearly simultaneous detector events comes from the same annihilation.
Random coincidences and scattered photons can contaminate PET data and must be estimated and corrected.
High sensitivity therefore does not mean every detected event is automatically valid signal.
18. Time-of-flight PET improves localisation along the response line
If detectors measure the tiny difference in arrival time between the two annihilation photons, reconstruction can estimate where along the line of response the event occurred.
This time-of-flight information improves signal localisation and noise behaviour.
It does not pinpoint the event exactly; timing resolution still defines an uncertainty range.
19. Reconstruction turns events into images
Modern PET and SPECT generally use iterative reconstruction algorithms that compare a candidate tracer distribution with measured detector data and repeatedly update the estimate.
System geometry, attenuation, scatter, detector response and noise models can be incorporated.
The final image is therefore a statistically reconstructed estimate of tracer distribution, not a raw detector photograph.
20. Spatial resolution changes apparent defect size
If a small perfusion region is near or below the scanner’s spatial resolution, its activity is averaged with neighbouring tissue.
This partial-volume effect can make small abnormalities look less intense and larger structures blur at their boundaries.
Apparent colour maps therefore inherit physical resolution limits even when reconstruction looks smooth.
21. Cardiac motion creates another reconstruction problem
The heart moves through systole and diastole and also shifts with breathing.
ECG gating can divide detected counts among cardiac phases, allowing gated SPECT or PET to estimate ventricular volumes and function.
But dividing counts into many gates reduces counts per gate, creating a trade-off between temporal information and statistical noise.
22. SPECT and PET answer overlapping but not identical measurement questions
SPECT remains widely available and has mature relative perfusion methods, while modern solid-state cameras are extending quantitative flow capability.
PET provides high sensitivity and established absolute myocardial blood-flow quantification with suitable tracers.
The 2025 review SPECT Myocardial Perfusion Imaging in the Era of PET discusses this evolving measurement landscape without making the two modalities interchangeable.
23. New SPECT systems can estimate myocardial blood flow
Cadmium-zinc-telluride cardiac cameras have much higher detection efficiency than conventional Anger-camera SPECT systems.
Dynamic acquisitions can therefore estimate tracer input and myocardial time-activity curves.
The review Myocardial blood flow quantification with SPECT summarises evidence and current repeatability limits.
24. Quantitative perfusion is becoming multimodality
PET is not the only modality capable of estimating absolute myocardial perfusion. CMR, CT and emerging echocardiographic techniques can also quantify flow under specified models.
The 2026 review Quantitative myocardial perfusion imaging across PET, SPECT, CMR, and CT compares this broader landscape.
Agreement across modalities depends on tracer physics, contrast kinetics, acquisition timing and modelling assumptions.
25. Worked problem: relative imaging can hide a global reduction
Suppose three myocardial territories receive stress flows of 1.2, 1.1 and 1.0 mL/min/g. If the highest region is normalised to 100%, the others appear only modestly lower.
Now compare a second heart with flows of 3.6, 3.3 and 3.0. The relative percentages are nearly the same even though absolute flows are three times larger.
Relative distribution and absolute perfusion answer different questions.
26. Worked problem: the same counts can mean different activity after attenuation
Region A and Region B produce the same detected counts, but photons from B pass through substantially more attenuating tissue.
After attenuation correction, B may be estimated to contain more tracer activity than A because a larger fraction of its photons was lost before detection.
Detector counts are therefore not identical to source activity.
27. Worked problem: flow reserve can change through the denominator
Model A has rest flow 1.0 and stress flow 3.0, giving reserve 3.0. Model B has rest flow 1.5 and stress flow 3.0, giving reserve 2.0.
Stress flow is identical, but the ratio differs because resting flow differs.
Flow reserve should therefore be interpreted together with both absolute numerator and denominator.
28. Worked problem: more counts are not always more spatial detail
Doubling acquisition time can improve counting statistics and reduce random image noise.
It does not halve detector crystal size or eliminate positron range and collimator resolution limits.
Signal-to-noise and spatial resolution are separate measurement properties.
29. The nuclear-perfusion mechanism in one causal chain
A radiotracer enters the bloodstream and reaches the myocardium according to coronary blood flow and tracer-specific extraction. Radioactive nuclei decay. In SPECT, emitted gamma photons pass through tissue and a collimator before detection. In PET, positron annihilation creates paired 511-keV photons detected in coincidence. Detected events are corrected for attenuation, scatter and system response and reconstructed into estimates of tracer distribution. Relative images compare myocardial regions, while dynamic tracer-kinetic modelling can convert time-activity curves into absolute myocardial blood flow. Repeating measurement at rest and during controlled hyperaemia yields flow reserve. Every final map or number therefore depends on coronary physiology, tracer kinetics, radiation physics, detector geometry and reconstruction mathematics.
Alicia stops calling nuclear perfusion a radioactive photograph. Tricia separates emitted photons from blood flow itself. Kai Kai writes the full chain: flow → tracer delivery → decay → photon detection → correction → reconstruction → physiological inference.
The deeper lesson is that nuclear imaging measures physiology through a labelled messenger. The scanner never sees blood flow directly; it observes radiation from molecules whose movement has been designed to reveal it.
Evidence trail and connected reading
For tracer principles, see the 2025 review Radionuclide Tracers for Myocardial Perfusion Imaging and Blood Flow Quantification. For current PET flow quantification, see Quantification of myocardial blood flow using PET. For SPECT developments, see SPECT Myocardial Perfusion Imaging in the Era of PET. The latest stress-imaging framework is the 2026 ASNC guideline.
Return to the parent: How the Heart Works. Continue to How Coronary Autoregulation Works, How Cardiac MRI Works, and How Cardiac CT Works.