Positron Emission Mammography (PEM) Physics

Positron emission mammography (PEM) is dedicated, high-resolution breast PET: two closely spaced planar detectors image F-18 FDG uptake in the mildly compressed breast at about 2.4 mm resolution — far finer than whole-body PET. By bringing the detectors to within a few centimeters of the breast, PEM converts the same positron physics used in whole-body FDG PET into a focused, small-field system optimized for depicting small, metabolically active breast lesions.^{1, 10} It is a coincidence (positron) technique — fundamentally different from molecular breast imaging with a cadmium-zinc-telluride (CZT) gamma camera, which images the single 140 keV photon of Tc-99m sestamibi — and different again from whole-body PET/CT staging and from receptor-targeted agents such as F-18 fluoroestradiol.⁹

This guide covers the instrumentation and detector physics of dedicated PEM, the close-geometry reason it out-resolves whole-body PET, a worked administered-activity and effective-dose calculation, the clinical evidence from the landmark PEM-versus-MRI trials, practical optimization, and the regulatory framework that governs any FDG imaging program. DRPS supports facilities running these systems through its PET/CT and nuclear medicine physics service.

Introduction

PEM answers a focused question: within this breast, where is FDG-avid disease, at a resolution fine enough to matter for surgical planning? Whole-body FDG PET has long been able to image breast cancer, but its 4 to 6 mm spatial resolution and the partial-volume effect blunt its ability to detect small primary tumors and ductal carcinoma in situ. PEM was developed to remove that geometric penalty by abandoning the whole-body ring and placing two small planar detectors directly above and below the gently compressed breast.^{1, 10}

The payoff is resolution. The commercial PEM Flex Solo II system — the device most of the clinical literature is built on — comprises two opposing detectors, each about 6 by 16.4 cm, that scan together to cover a field of view up to 24 by 16.4 cm, and achieves an in-plane spatial resolution of about 2.4 mm in full width at half maximum.^{1, 12} That is a real improvement over whole-body PET and lowers the threshold on lesion size and tracer uptake for detection within the breast.¹

PEM is not a replacement for mammography, for whole-body staging, or for breast MRI. It is a dedicated, problem-solving and local-staging tool. Understanding its physics — why it resolves better, what it gives up in exchange, and how much dose it costs — is what separates appropriate use from overuse.

Topic Explanation

What PEM is, physically

PEM is coincidence imaging of F-18 FDG with two close, opposed planar detectors. FDG is a glucose analog labeled with fluorine-18. After intravenous injection and an uptake period, FDG concentrates in metabolically active tissue. Fluorine-18 decays by positron emission; each emitted positron annihilates with a nearby electron to produce two 511 keV photons traveling in nearly opposite directions. A PEM scanner detects these photon pairs in coincidence — just as a whole-body PET scanner does — but with the detectors brought close to the breast rather than arranged in a meter-wide ring.^{1, 9}

The breast is placed under mild compression (far gentler than screening mammography) between the two detector heads. Limited-angle tomographic reconstruction then produces a stack of image planes parallel to the detector faces. Because the two heads do not surround the breast, PEM samples only a limited range of angles, which is the root of several of its physics characteristics discussed below.^{1, 12}

How PEM differs from its neighbors

PEM is frequently confused with other functional breast techniques. The distinctions are physical and clinically important:

• Molecular breast imaging (MBI) / breast-specific gamma imaging (BSGI) uses a CZT gamma camera to image the single 140 keV photon of Tc-99m sestamibi — a single-photon modality. PEM is a coincidence modality imaging two 511 keV photons from F-18 FDG. For the single-photon approach, see our guide to molecular breast imaging with CZT.⁹
• Whole-body PET/CT images the whole body with a large detector ring and an integrated CT for attenuation correction and localization; it stages nodal and distant disease but resolves the breast more coarsely.¹⁰
• F-18 fluoroestradiol (FES) PET images estrogen-receptor expression rather than glucose metabolism; see F-18 fluoroestradiol (FES) PET for breast cancer.

The instrument

The detectors in the clinical PEM system are lutetium-based scintillators (LYSO) read out to localize each 511 keV interaction, cleared by the FDA under 510(k) K090553 as a high-resolution, small-field-of-view PET imaging system.¹² A companion stereotactic accessory adds PEM-guided, vacuum-assisted biopsy, so a lesion visible only on PEM can be sampled under positron guidance.⁷ The system reports semiquantitative uptake metrics analogous to the standardized uptake value used across PET, although — as detailed below — PEM quantification is limited by the geometry.¹

Key Technical Principles

Why close geometry resolves better

Spatial resolution in any PET system is the quadrature sum of several blurring terms: the positron range, the 180-degree annihilation photon non-collinearity, the detector element size, and reconstruction effects. Two of these favor a dedicated breast system.

First, F-18 is intrinsically well suited to high-resolution imaging. Its positrons have a low maximum energy (about 0.63 MeV) and therefore a short range in tissue before annihilation, contributing only a few tenths of a millimeter of blur — much less than higher-energy emitters.¹

Second, and decisively for PEM, the non-collinearity term scales with the separation between the detectors. The two annihilation photons depart not exactly 180 degrees apart but with a small angular spread of roughly 0.5 degrees FWHM. The resulting positional blur is approximately:

where ${\Delta }_{\text{nc}}$ is the non-collinearity blur (FWHM) and $D$ is the distance between the two detectors (same units). For a whole-body ring with $D\approx 80\text{cm}$:

For a dedicated PEM geometry with the heads a compressed-breast thickness apart, $D\approx 6\text{cm}$:

The non-collinearity penalty essentially vanishes at close range. Combined with the detectors' proximity raising geometric count sensitivity, this is the physical reason a dedicated PEM camera delivers roughly 2.4 mm in-plane resolution where whole-body PET delivers 4 to 6 mm.^{1, 10} The 2.4 mm FWHM figure was measured on the clinical system for image planes parallel to the detector faces.¹

Administered activity and effective dose (worked example)

PEM uses F-18 FDG at a labeled administered activity of 370 MBq (10 mCi).¹⁰ The effective dose follows from the ICRP Publication 128 dose coefficient for FDG, $e\approx 0.019\text{mSv/MBq}$:¹³

where $E$ is the effective dose, $A$ is the administered activity, and $e$ is the FDG effective dose coefficient. A single PEM study at the labeled activity therefore delivers an effective dose on the order of 7 mSv — whole-body, because FDG distributes systemically — which is why a PEM study carries a radiation-induced-cancer risk far higher than a two-view digital mammogram.^{10, 11} One analysis estimated the lifetime attributable risk of fatal cancer from a single 370 MBq PEM study to be roughly 23 times that of digital mammography in a 40-year-old woman.¹⁰ This dose accounting is the single strongest argument for restricting PEM to appropriate diagnostic indications and for pursuing reduced-activity protocols rather than using it for routine screening.¹¹

Note also the decay constraint: F-18 has a physical half-life of about 110 minutes, so the dose must be calibrated to the administration time and the acquisition scheduled against decay, exactly as for any FDG study. For the detailed dose-reduction framework, see F-18 FDG PET dose optimization.

What the geometry costs

The same limited-angle geometry that buys resolution imposes limits that a physicist must understand and communicate:

Measured background (uniformity) variability was about 5 percent, positioning linearity began roughly 13 mm from the edge of the detector housing, and the coefficient of variation rose sharply at the chest-wall edge of the field of view where sampling is sparse.¹ These are not defects but intrinsic consequences of the dedicated geometry, and they define where PEM is reliable and where it is not.

PEM versus the alternatives

The MBI activity shown is the label-recommended Tc-99m sestamibi dose used in the dose-risk analysis; modern low-dose MBI protocols use substantially less. The point of the table is not a winner but a map: PEM trades whole-body coverage for resolution, trades MRI's lesion sensitivity for higher specificity, and carries a non-trivial systemic radiation dose that MRI does not.^{3, 9, 10}

Clinical Impact

Primary breast cancer detection

In the breast with known or suspected cancer, PEM is accurate. A prospective multicenter study of high-resolution FDG PET with compression reported sensitivity of about 90 percent, specificity of 86 percent, positive and negative predictive values of 88 percent, and an area under the ROC curve of 0.918; it depicted 10 of 11 cases of ductal carcinoma in situ (91 percent) and 33 of 37 invasive cancers (89 percent), including subcentimeter tumors.² In a dedicated reader study, trained breast imagers interpreting PEM achieved mean sensitivity of 96 percent and specificity of 84 percent with an area under the curve of 0.95 — higher than the same readers' performance on MRI in that task.⁵

PEM versus MRI for extent of disease

The pivotal comparison came from a 472-woman multicenter trial of PEM versus contrast-enhanced MRI in women with newly diagnosed cancer. For the ipsilateral (same-side) breast, PEM and MRI had comparable breast-level sensitivity; MRI had greater lesion-level sensitivity and more accurately predicted the need for mastectomy, while PEM had greater specificity at both the breast and lesion levels, and the positive predictive value of PEM-prompted biopsy (66 percent) exceeded that of MRI-prompted biopsy (53 percent).³ Combining PEM and MRI detected more additional cancer than MRI alone.³ A standardized PEM lexicon — patterned on BI-RADS — was developed and validated so that findings could be reported and managed consistently.^{4, 5}

Where PEM underperforms

The same trial exposed PEM's main weakness. For screening the contralateral (opposite) breast, MRI found 14 of 15 cancers (93 percent) while PEM prospectively identified only 3 of 15 (20 percent).⁸ The lesson built into current practice is twofold: PEM is a tool for the breast you are already worried about, not a general screening system; and a lesion visible on PEM should be treated as suspicious unless already known to be benign.^{6, 8}

Biopsy guidance

Because PEM can show disease not seen on other modalities, PEM-guided biopsy closes the loop. In a prospective multicenter study, 24 PEM-guided vacuum-assisted biopsies all successfully targeted and sampled their lesions, invasive cancer was found in 13 of 24 (54 percent), and the procedure was safe with only minimal to mild discomfort.⁷

Practical Optimization Tips

• Reserve PEM for defined indications. Its ~7 mSv systemic dose makes it inappropriate for routine screening; use it for local staging, extent-of-disease, and problem-solving in the breast with known or suspected cancer.^{10, 11}
• Pursue reduced-activity protocols where validated. Dose scales directly with administered activity, so any defensible reduction below the labeled 370 MBq lowers risk proportionally.¹¹
• Control the uptake period and glucose. As with any FDG study, standardize the uptake time and confirm acceptable blood glucose so uptake and semiquantitative values are comparable.
• Position to capture the chest wall, but read the edge cautiously. Sensitivity and quantification degrade at the chest-wall edge of the field of view; know where the reliable field ends.¹
• Treat PEM numbers as semiquantitative. Without full attenuation and scatter correction, uptake values vary within the field of view; do not use them as if they were fully corrected SUVs.¹
• Use the standardized lexicon. Report with the validated PEM feature lexicon so assessments and malignancy rates are consistent across readers.^{4, 5}
• Calibrate the dose calibrator and synchronize clocks. Accurate administered activity and decay correction depend on a quality-controlled dose calibrator and synchronized injection and acquisition times.
• Pair PEM with the right partner study. Its strength is specificity in the ipsilateral breast; pair it with MRI or whole-body imaging when lesion-level sensitivity or distant staging is the question.^{3, 8}

Regulatory Considerations

PEM sits at the intersection of FDA device regulation and NRC or Agreement State control of the radioactive drug. The scanner itself is an FDA-cleared medical device; the F-18 FDG it images is byproduct material whose possession and medical use are licensed and inspected under radiation-control regulations.^{12, 14}

Key frameworks to align with:

• FDA 510(k) clearance (K090553) — the PEM Flex Solo II was cleared as a high-resolution, small-field-of-view PET imaging system; the device must be used within its cleared intended use.¹²
• 10 CFR Part 35, Medical Use of Byproduct Material — governs authorized users, dose calibrator quality control, and radiation safety for diagnostic PET radiopharmaceuticals such as FDG. Agreement States administer equivalent programs.¹⁴
• ICRP Publication 128 — the reference compendium for radiopharmaceutical dose coefficients used to justify and document patient dose.¹³
• ACR PET/CT accreditation and practice parameters — scanner performance, dose calibrator QC, and qualified-medical-physicist oversight that any FDG imaging program is expected to meet.

Across the states DRPS serves, Florida, Maryland, Virginia, California, Nevada, Pennsylvania, New York, and New Jersey are NRC Agreement States that license the medical use of byproduct material under their own radiation-control programs, while Washington, DC and Delaware are regulated directly by the NRC. A facility must confirm which authority issues its license and what authorized-user, QC, and reporting requirements apply before imaging patients with FDG. For related performance context, see PET detector scintillators and SiPMs and PET spatial resolution and positron range.

Frequently Asked Questions (FAQs)

What is positron emission mammography (PEM)?

PEM is dedicated, high-resolution breast positron emission tomography. After an intravenous injection of F-18 fluorodeoxyglucose (FDG), the mildly compressed breast is imaged between two closely spaced planar detectors that record the paired 511 keV annihilation photons from positron decay. Because the detectors sit close to the breast, PEM achieves an in-plane spatial resolution of about 2.4 mm — substantially finer than whole-body PET — and so can depict small, FDG-avid lesions within the breast.

How is PEM different from molecular breast imaging (MBI) with CZT?

They use different physics. PEM is a coincidence (positron) technique: it detects the two opposed 511 keV photons from F-18 FDG annihilation. Molecular breast imaging with a cadmium-zinc-telluride (CZT) gamma camera is a single-photon technique that images the 140 keV gamma ray from Tc-99m sestamibi. PEM images glucose metabolism with F-18 FDG; MBI images sestamibi uptake. The detector hardware, the radiopharmaceutical, and the energy of the detected photons are all distinct.

How is PEM different from whole-body PET/CT?

PEM uses two small planar detectors placed close to the breast instead of a large-diameter detector ring around the body. The close geometry and small detector elements give PEM roughly 2.4 mm in-plane resolution versus the 4 to 6 mm typical of whole-body PET, improving detection of small breast lesions. The trade-off is a limited field of view confined to the breast and limited-angle tomography, so PEM does not stage the axilla or distant disease the way whole-body PET/CT does.

What radiopharmaceutical and activity does PEM use?

PEM uses F-18 FDG, the same glucose analog used for whole-body FDG PET. The labeled administered activity reported in the dosimetry literature is 370 MBq (10 mCi). Using the ICRP Publication 128 effective dose coefficient for FDG of about 0.019 mSv/MBq, that corresponds to an effective dose of roughly 7 mSv, which is why dose justification and low-activity protocols are an active part of PEM practice.

How accurate is PEM for breast cancer?

In a prospective multicenter study of primary breast cancer, PEM had about 90 percent sensitivity and 86 percent specificity with an area under the ROC curve of 0.918, and it depicted the majority of ductal carcinoma in situ and invasive cancers. In a reader study, trained breast imagers achieved mean sensitivity of 96 percent and specificity of 84 percent. Performance is strongest for evaluating the breast with known or suspected cancer; PEM was far less sensitive than MRI for screening the opposite (contralateral) breast.

Can PEM guide a biopsy?

Yes. A PEM-guided, vacuum-assisted biopsy accessory lets a lesion seen only on PEM be sampled under positron-emission guidance. In a prospective multicenter study, all targeted lesions were successfully sampled and invasive cancer was found in about half, establishing PEM-guided biopsy as a safe and effective way to work up a PEM-only finding.

What are the main physics limitations of PEM?

PEM uses limited-angle tomography with two opposed detectors rather than a full ring, so depth (cross-plane) resolution is poorer than in-plane resolution and quantification is not fully corrected for attenuation and scatter. Sensitivity falls near the chest wall at the edge of the field of view, where sampling is limited, and the field of view is confined to the breast. These are the reasons PEM is a dedicated problem-solving and local-staging tool rather than a whole-body or standalone screening system.

Key Takeaways

• PEM is coincidence breast PET. It images F-18 FDG with two close, opposed planar detectors — distinct from single-photon MBI with CZT and Tc-99m sestamibi, and from whole-body PET/CT.
• Close geometry buys resolution. Bringing the detectors to within a few centimeters of the breast collapses the non-collinearity blur and delivers about 2.4 mm in-plane resolution, versus 4 to 6 mm for whole-body PET.
• The dose is systemic and non-trivial. At the labeled 370 MBq of FDG, the effective dose is roughly 7 mSv — far more than a mammogram — which rules out routine screening and motivates reduced-activity protocols.
• It is accurate in the at-risk breast. Prospective data show about 90 percent sensitivity and 86 percent specificity for primary cancer, with greater specificity than MRI in the ipsilateral breast.
• It is not a screening tool. PEM was far less sensitive than MRI for the contralateral breast; its role is local staging and problem-solving, not whole-body or bilateral screening.
• Geometry sets the limits. Limited-angle sampling means poorer depth resolution, semiquantitative values, and reduced reliability at the chest-wall edge.

Conclusion

Positron emission mammography took the well-understood physics of FDG PET and re-engineered the geometry for one organ. By moving the detectors close to the breast, PEM nearly eliminates the non-collinearity blur and raises count sensitivity, achieving a resolution that whole-body PET cannot match within the breast — and the clinical trials bear that out, with strong specificity and accuracy in the breast that already carries a diagnosis.

But the same physics that defines its strengths defines its boundaries. Limited-angle reconstruction leaves depth resolution and quantification imperfect, the field of view stops at the breast, and the roughly 7 mSv systemic dose of an FDG injection keeps PEM firmly in the diagnostic and staging domain rather than the screening domain. Used for the questions it was built to answer — extent of disease, problem-solving, and biopsy guidance in the breast with known or suspected cancer — PEM is a precise, evidence-supported tool. The physicist's job is to keep it there: correct activity and dose justification, honest quantification, and quality control that matches the geometry.

How DRPS Can Help

Diagnostic Radiation Physics Services supports PET/CT and nuclear medicine facilities operating dedicated and whole-body molecular breast imaging — with PET/CT and nuclear medicine physics, dose calibrator quality control, administered-activity and effective-dose justification, scanner performance evaluation, mammography and breast-imaging physics, and board-certified medical physicist consulting. All work is performed by board-certified medical physicists.

DRPS supports facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, Pennsylvania, New York, New Jersey, and Delaware.

A dedicated breast PET program is only as good as the dose justification and quality control behind it — and that is exactly what a qualified medical physicist brings.

Related Resources

• Molecular breast imaging with CZT
• F-18 fluoroestradiol (FES) PET for breast cancer
• PET spatial resolution and positron range
• F-18 FDG PET dose optimization
• PET detector scintillators and SiPMs
• PET/CT and nuclear medicine physics
• Mammography physics and MQSA
• Medical physicist consulting

References

1. MacDonald L, Edwards J, Lewellen T, Haseley D, Rogers J, Kinahan P. Clinical imaging characteristics of the positron emission mammography camera: PEM Flex Solo II. Journal of Nuclear Medicine. 2009;50(10):1666-1675. doi:10.2967/jnumed.109.064345. PubMed
2. Berg WA, Weinberg IN, Narayanan D, et al. High-resolution fluorodeoxyglucose positron emission tomography with compression ("positron emission mammography") is highly accurate in depicting primary breast cancer. The Breast Journal. 2006;12(4):309-323. doi:10.1111/j.1075-122X.2006.00269.x. PubMed
3. Berg WA, Madsen KS, Schilling K, et al. Breast cancer: comparative effectiveness of positron emission mammography and MR imaging in presurgical planning for the ipsilateral breast. Radiology. 2010;258(1):59-72. doi:10.1148/radiol.10100454. PubMed
4. Narayanan D, Madsen KS, Kalinyak JE, Berg WA. Interpretation of positron emission mammography: feature analysis and rates of malignancy. AJR American Journal of Roentgenology. 2011;196(4):956-970. doi:10.2214/AJR.10.4748. PubMed
5. Narayanan D, Madsen KS, Kalinyak JE, Berg WA. Interpretation of positron emission mammography and MRI by experienced breast imaging radiologists: performance and observer reproducibility. AJR American Journal of Roentgenology. 2011;196(4):971-981. doi:10.2214/AJR.10.5081. PubMed
6. Berg WA. Nuclear breast imaging: clinical results and future directions. Journal of Nuclear Medicine. 2016;57(Suppl 1):46S-52S. doi:10.2967/jnumed.115.157891. PubMed
7. Kalinyak JE, Schilling K, Berg WA, et al. PET-guided breast biopsy. The Breast Journal. 2011;17(2):143-151. doi:10.1111/j.1524-4741.2010.01044.x. PubMed
8. Berg WA, Madsen KS, Schilling K, et al. Comparative effectiveness of positron emission mammography and MRI in the contralateral breast of women with newly diagnosed breast cancer. AJR American Journal of Roentgenology. 2012;198(1):219-232. doi:10.2214/AJR.10.6342. PubMed
9. Narayanan D, Berg WA. Dedicated breast gamma camera imaging and breast PET: current status and future directions. PET Clinics. 2018;13(3):363-381. doi:10.1016/j.cpet.2018.02.008. PubMed
10. Hendrick RE. Radiation doses and cancer risks from breast imaging studies. Radiology. 2010;257(1):246-253. doi:10.1148/radiol.10100570. PubMed
11. Hendrick RE. Radiation doses and risks in breast screening. Journal of Breast Imaging. 2020;2(3):188-200. doi:10.1093/jbi/wbaa016. PubMed
12. U.S. Food and Drug Administration. 510(k) Premarket Notification K090553: PEMFlex Solo II High Resolution PET Scanner. Cleared March 31, 2009. fda.gov
13. International Commission on Radiological Protection. ICRP Publication 128: Radiation Dose to Patients from Radiopharmaceuticals — A Compendium of Current Information Related to Frequency and Dose. Annals of the ICRP. 2015;44(2S). icrp.org
14. U.S. Nuclear Regulatory Commission. 10 CFR Part 35: Medical Use of Byproduct Material. ecfr.gov