# PET/CT and SPECT/CT: Hybrid Imaging Principles

## Rationale for Hybrid Imaging

### Why Combine Functional and Anatomic Imaging

Nuclear medicine images excel at revealing functional and molecular processes but inherently lack the anatomic detail needed to precisely localize abnormalities. CT, on the other hand, provides superb anatomic resolution but limited functional information. Hybrid imaging co-registers these complementary datasets, allowing clinicians to pinpoint exactly where a functional abnormality sits within the body's anatomy. This fusion improves diagnostic confidence, sensitivity, and specificity compared to interpreting either modality in isolation. From the patient's perspective, a single-session acquisition is more convenient and efficient than scheduling separate studies.

### Historical Development

The first PET/CT prototype was developed by Townsend and Beyer in the late 1990s, and the concept rapidly proved its clinical value. SPECT/CT followed in the early 2000s, initially with low-dose CT systems like the Hawkeye. By the mid-2000s, PET/CT had become the standard of care for oncologic imaging. Today, virtually all new PET systems and an increasing proportion of SPECT systems are hybrid devices incorporating CT capability.

## CT Attenuation Correction

### Principles

Both PET photons (511 keV) and SPECT photons (140 keV for Tc-99m) are attenuated as they traverse tissue. Without correction, structures deep within the body appear artificially cold while peripheral structures appear relatively hot. The CT component of a hybrid scanner provides a high-resolution map of the patient's internal density, which can be converted into an attenuation correction map. CT Hounsfield units are transformed into linear attenuation coefficients at the appropriate photon energy using bilinear scaling algorithms.

### PET/CT Attenuation Correction

For PET, the CT values acquired at an effective energy of roughly 70 to 80 keV must be scaled to the 511 keV annihilation photon energy. The bilinear conversion uses one slope for soft tissue densities (below approximately 0 HU) and a different slope for bone densities (above 0 HU). A low-dose CT scan at 10 to 50 mAs is sufficient for attenuation correction purposes, though a full diagnostic CT can optionally be performed. CT-based attenuation correction is faster and produces less noisy correction maps than the older method of using external Ge-68 transmission rod sources.

### SPECT/CT Attenuation Correction

The same principle applies for SPECT, with the CT values scaled to the appropriate gamma ray energy. Earlier SPECT/CT systems used low-resolution, slow CT scanners such as the Hawkeye flat-panel system, while modern systems incorporate diagnostic-quality multi-slice CT scanners. In cardiac SPECT, CT-based attenuation correction substantially reduces the soft-tissue artifacts that commonly affect women (breast attenuation of the anterior wall) and men (diaphragmatic attenuation of the inferior wall).

## Image Co-Registration

### Hardware Co-Registration

In hybrid scanners, the patient remains on the same table for both the CT and nuclear acquisitions. The CT and PET or SPECT gantries share a common coordinate system, and acquisition proceeds sequentially: the CT scan is completed first (in seconds to minutes), followed by the nuclear emission scan (in minutes to tens of minutes). This hardware co-registration provides reliable spatial alignment as long as the patient remains still between acquisitions.

### Software Fusion

When images are acquired on separate scanners, software-based registration algorithms can retrospectively align the datasets using rigid or non-rigid transformations. However, this approach is inherently less reliable than hardware co-registration because of differences in patient positioning between scans. Software fusion is used when hybrid scanners are unavailable or when correlating nuclear medicine findings with previously acquired anatomic imaging.

## Artifacts in Hybrid Imaging

### Respiratory Motion Misregistration

This is the most frequently encountered artifact in hybrid imaging. The CT scan is acquired during a single breath-hold or quiet breathing over just seconds, capturing the anatomy at one moment in the respiratory cycle. The PET or SPECT scan, by contrast, is acquired over several minutes of free breathing, producing an image that averages over many respiratory cycles. This temporal mismatch causes spatial misregistration at the lung base and diaphragm, which can produce a curvilinear cold artifact at the dome of the liver on PET, apparent focal uptake at the lung base where shifted lung activity is projected onto the liver on CT, and mislocalization of lesions near the diaphragm. Mitigation strategies include respiratory gating, shallow breathing CT protocols, and cine CT averaging.

### Metallic Implant Artifacts

Dense metallic objects such as hip prostheses, dental hardware, and implanted ports produce very high CT attenuation values. When these elevated values are used for attenuation correction, the algorithm overestimates the photon attenuation at these sites, creating false-positive hot spots on the corrected PET or SPECT images. The key to recognizing this artifact is reviewing the non-attenuation-corrected (NAC) images: if the apparent uptake disappears on NAC images while coinciding with metallic hardware on CT, it is an artifact.

### Contrast Artifacts

Intravenous contrast can elevate CT attenuation values and lead to overcorrection on PET or SPECT. This is most significant in areas of concentrated contrast such as the superior vena cava, subclavian veins, and kidneys during the bolus phase. Oral contrast can also cause artifacts, with barium-based preparations being more problematic than water-based contrast. In practice, the clinical impact is usually minor, but awareness is important. Some centers acquire a dedicated low-dose CT for attenuation correction separate from the diagnostic contrast-enhanced CT.

### Truncation Artifacts

When the patient extends beyond the CT field of view, such as when the arms are positioned at the sides, the CT data are incomplete at the periphery. This truncated data produces inaccurate attenuation correction in the affected regions, typically manifesting as a bright rim artifact at the edges of the CT field. Extended field-of-view reconstruction algorithms can mitigate this problem.

### Patient Motion

Any movement between the CT and nuclear acquisitions causes spatial misregistration, leading to both incorrect attenuation correction and inaccurate anatomic localization of functional findings. Motion is most common in head-and-neck studies (due to swallowing), extremity imaging, and lengthy acquisition protocols. Reviewing the fused images for alignment before interpretation is an essential quality control step.

| Artifact | Cause | Appearance on AC Images | Key to Recognition | Mitigation |
|---|---|---|---|---|
| Respiratory misregistration | CT vs emission temporal mismatch | Curvilinear cold band at liver dome | Absent on NAC images; check fusion | Respiratory gating, shallow-breathing CT |
| Metallic implant | High-density overcorrection | False hot spot at implant | Absent on NAC; correlate with CT | Review NAC images |
| IV contrast | Concentrated contrast overcorrection | Apparent uptake in SVC/subclavian | Absent on NAC; bolus phase on CT | Separate low-dose CTAC scan |
| Truncation | Patient beyond CT FOV | Bright rim at periphery | Edge artifact on CT | Extended FOV reconstruction |
| Patient motion | Movement between scans | Mislocalized uptake | Misalignment on fused images | Repeat scan, immobilization |

## Low-Dose CT vs. Diagnostic CT

### Low-Dose CT

Low-dose CT protocols typically use 10 to 50 mAs at 100 to 120 kVp. Their purpose is limited to attenuation correction and anatomic localization, and they are not considered diagnostic quality for primary anatomic interpretation. The radiation dose from a low-dose CT is modest, approximately 1 to 3 mSv, considerably lower than a full diagnostic CT.

### Diagnostic CT Protocol

A full-dose, contrast-enhanced CT with dedicated organ-specific protocols provides both CTAC data and a fully diagnostic anatomic study. This approach can eliminate the need for a separate diagnostic CT appointment, improving efficiency and reducing inconvenience. However, it increases the total radiation dose and requires coordination with radiology for protocol optimization, intravenous contrast administration, and timing.

### Controversy

There is ongoing institutional debate about the routine use of diagnostic-quality CT with PET. Proponents argue it eliminates redundant imaging and improves staging accuracy. Opponents cite increased radiation dose, the complexity of contrast administration, and concerns about overdiagnosis of incidental findings. Practice varies widely, with some centers routinely performing diagnostic CT and others using low-dose CTAC-only protocols.

## Clinical Impact of Hybrid Imaging

### Oncology

PET/CT has transformed oncologic staging, changing the TNM classification in 25 to 30% of cancer patients compared to PET alone or CT alone. By precisely localizing FDG-avid foci to specific lymph node stations, differentiating bone from soft tissue lesions, and distinguishing tumor from physiologic muscle uptake, hybrid imaging guides biopsy site selection and improves radiation therapy planning.

### Cardiology

SPECT/CT with CT-based attenuation correction reduces the attenuation artifacts that have historically plagued myocardial perfusion imaging. The CT component can also provide coronary calcium scoring, and combined PET/CT protocols can integrate perfusion imaging with coronary CT angiography in selected patients.

### Musculoskeletal

SPECT/CT has dramatically improved the specificity of bone scintigraphy. When increased bone uptake is identified on the nuclear component, the co-registered CT provides morphologic characterization, distinguishing degenerative changes from fractures from metastatic disease. This combined approach is particularly valuable for evaluating the spine, foot, and knee.

<image>A schematic diagram showing the layout of a PET/CT scanner with the CT gantry and PET detector ring on a shared patient bed. Illustrate the sequential acquisition process: the patient first passes through the CT gantry (fast scan, seconds), then through the PET ring (slower scan, minutes per bed position). Show how the shared coordinate system enables precise hardware co-registration of the CT anatomic map with the PET functional data, producing fused images.</image>

<image>A series of clinical PET/CT images demonstrating common artifacts. Panel A: respiratory motion misregistration showing curvilinear cold artifact at the liver dome with apparent focal uptake at the lung base. Panel B: metallic hip prosthesis causing false-positive hot spot on attenuation-corrected PET images that disappears on non-attenuation-corrected images. Panel C: IV contrast artifact with apparent increased activity in the subclavian vein region on CTAC PET. Panel D: truncation artifact with bright rim at the periphery where patient arms extend beyond the CT field of view.</image>

<image>A side-by-side comparison of myocardial perfusion SPECT without and with CT-based attenuation correction. Show a male patient with apparent inferior wall perfusion defect on non-corrected images (diaphragmatic attenuation artifact) that normalizes after CTAC, and a female patient with apparent anterior wall defect (breast attenuation artifact) that resolves with CTAC. Annotate the attenuation maps showing soft tissue causing the artifacts.</image>

## Clinical Pearls

Always review non-attenuation-corrected images alongside CTAC images. Any focal uptake that appears only on the attenuation-corrected images and is absent on the NAC images should raise suspicion for an attenuation correction artifact, particularly when it coincides with high-density material on CT.

Respiratory misregistration is the most common artifact at the lung base and diaphragm. Its classic appearance is a curvilinear photopenic band at the liver dome on PET, sometimes accompanied by apparent focal uptake at the lung base.

Metallic implants create false hot spots on CTAC PET images because the algorithm overestimates attenuation at metal densities. The NAC images will reliably demonstrate the absence of true uptake.

When interpreting PET/CT, always verify spatial alignment between the CT and nuclear components on the fused images before beginning interpretation. Misregistration degrades both quantitative accuracy and anatomic localization.

CT-based attenuation correction in cardiac SPECT significantly reduces soft-tissue artifacts and improves diagnostic specificity, particularly for the inferior wall in men and the anterior wall in women.

Low-dose CT for CTAC adds only 1 to 3 mSv of radiation dose, while a full diagnostic CT adds 5 to 15 mSv. The benefit of diagnostic-quality CT must be weighed against this additional radiation exposure on a case-by-case basis.

## References

- Beyer T, et al. A combined PET/CT scanner for clinical oncology. *J Nucl Med*. 2000;41(8):1369-1379.
- Bockisch A, et al. Hybrid imaging by SPECT/CT and PET/CT. *J Nucl Med*. 2009;50(Suppl 1):24S-34S.
- Kinahan PE, et al. Attenuation correction for a combined 3D PET/CT scanner. *Med Phys*. 1998;25(10):2046-2053.
- Delbeke D, et al. Procedure guideline for SPECT/CT imaging 1.0. *J Nucl Med*. 2006;47(7):1227-1234.