PET

Principle

Positron emission tomography (PET) is a nuclear medicine medical imaging technique which produces a three-dimensional image or map of functional processes in the body. PET is both a medical and research tool.

Fig. 1 Schematic view of a detector block and ring of a PET scanner.

To conduct the scan, a short-lived radioactive tracer isotope (radionuclide), which decays by emitting a positron (positive electron, the antimatter counterpart of an electron), which also has been chemically incorporated into a metabolically active molecule (the radiotracers), is injected into the living subject (usually into blood of a patient). There is a waiting period while the metabolically active molecule becomes concentrated in tissues of interest; then the research subject or patient is placed in the imaging scanner. The molecule most commonly used for this purpose is the sugar F¹⁸-deoxyglucose (FDG). Radionuclides used in PET scanning are typically isotopes with short half lives such as C¹¹ (20 min), N¹³ (10 min), O¹⁵ (2 min), and F¹⁸ (110 min). Due to their short half lives, the radionuclides must be produced in a nearby cyclotron (a circular charged-particle accelerator).

As the radioisotope undergoes positron emission decay (also known as positive beta decay), it emits a positron. After annihilation (see Gamma camera) two (sometimes more) gamma) photons are produced moving in almost opposite directions. These are detected when they reach a scintillator (see Gamma camera) material in the scanning device, creating a burst of light which is detected by photomultiplier tubes or nowadays often with silicon photodiodes (Si APD). The technique depends on simultaneous or coincident detection of the pair of photons. Photons which do not arrive in pairs (i.e., within a few nanoseconds) are ignored.

Fig. 2 Schema of a PET acquisition process

The PET scanner examines detailed molecular biological functions via the use of radiolabelled molecular probes that have different rates of uptake, depending on the type and function of tissue involved. The changing of regional blood flow in various anatomic structures (as a measure of the injected positron emitter) can be visualized and relatively quantified with a PET scan.

PET imaging is best performed using a dedicated PET scanner. However, it is possible to acquire PET images using a conventional dual-head Gamma camera fitted with a coincidence detector (lower quality, slower).

Image reconstruction The raw data are a list of the annihilation photons by a pair of detectors. Each coincidence event represents a line in space connecting the two detectors along which the positron emission occurred.

Coincidence events can be grouped into projections images, called sinograms. The sinograms are sorted by the angle of each view and tilt, the latter in 3D case images. The sinogram images are analogous to the projections captured by CT scanners, and can be reconstructed in a similar way. However, the statistics of the data is much worse than those obtained through transmission tomography. A normal PET data set has millions of counts for the whole acquisition, while the CT can reach a few billion counts. Furthermore, PET data suffer from scatter and random events much more dramatically than CT data does.

In practice, considerable pre-processing of the data is required - correction for random coincidences, estimation and subtraction of scattered photons, detector dead-time correction (after the detection of a photon, the detector must "cool down" again) and detector-sensitivity correction (for both inherent detector sensitivity and changes in sensitivity due to angle of incidence). For more details see More Info.

Fig. 3 Maximum intensity projection (MIP, see Image processing: 3D reconstruction) of a typical ¹⁸FDG nearly whole body PET acquisition.

Limitations Limitations of PET are ethical since radioactive material is injected. Most radionuclides requires a cyclotron and automated chemistry lab for radiopharmaceutical manufacture in the hospital. This limitation restricts clinical PET primarily to the use of tracers labeled with ¹⁸F or ⁸²Rb, due to their longer half lifes. The latter, used for myocardial perfusion studies. Mobile cyclotrons with hot labs are a new development, meeting the cost problems.

Safety PET scanning is non-invasive, but it does involve exposure to ionizing radiation. The total dose of radiation is small, however, usually around 7 mSv. This can be compared to about 2.2 mSv average annual background radiation, 0.02 mSv for a chest x-ray, up to 8 mSv for a CT scan of the chest and 2-6 mSv per annum for aircrew (data from UK National Radiological Protection Board).

Application

PET is a valuable technique for some diseases and disorders, because it is possible to target the radio-chemicals used for particular functions and processes.

Oncology FDG-PET scanning with the glucose analog ¹⁸FDG with a typical dose of 200-400 MBq. It results in intense radiolabeling of tissues with high glucose uptake, such as the brain, the liver, and most cancers.

Neurology/neuroscience PET neuro-imaging is based on areas of higher radioactivity. What is actually measured indirectly is the flow of blood to different parts of the brain by an ¹⁵O tracer. However, because of its 2-minute half-life ¹⁵O must be piped directly from a medical cyclotron, and this is difficult and expensive. Therefore, standard FDG-PET may also be successfully used, e.g. to differentiate Alzheimer's disease (recently via visualization of amyloid plaques) from other dementing processes.

Specific radiotracers (i.e. radioligands) have been developed for PET that are ligands for neuroreceptor subtypes (for e.g. dopamine D2, serotonin 5-HT1A, etc.), and in addition transporters (such as for serotonin in this case), or enzyme substrates. These agents permit the visualization of neuroreceptor pools in the context of a plurality of neuropsychiatric and neurological illnesses.

Psychiatry Numerous compounds that bind selectively to neuroreceptors have been radiolabeled with ¹¹C or ¹⁸F. Radioligands that bind to dopamine receptors (D1,D2, reuptake transporter), serotonin receptors (5HT1A, 5HT2A, reuptake transporter) opioid receptors (mu) and other sites have been used successfully in studies with human subjects.

Pharmacology In pre-clinical trials, it is possible to radiolabel a new drug, inject it into animals and monitored in vivo.

Cardiovascular system So-called "hibernating myocardium", and imaging atherosclerosis to detect patients at risk of stroke.

PET is increasingly used together with CT or magnetic resonance imaging (MRI), the latter giving anatomic information. Modern PET scanners are integrated high-end multi-detector-row CT scanners. The two scans can be performed during the same session, with the patient not changing position.

Fig. 3 PET scan of the human brain.

More Info

The most significant fraction of electron-positron decays result in two 511 keV gamma photons being emitted at almost 180 ^o to each other; the "line of response" or LOR. In practice the LOR has a finite width as the emitted photons are not exactly 180^o apart. If the recovery time of detectors is in the picosecond range rather than the 10's of ns range, it is possible to calculate the single point on the LOR at which an annihilation event originated, by measuring the "time of flight" of the two photons. This technology is not yet common, but it is available on some new systems. More commonly, a technique very like the reconstruction of CT and SPECT data is used, although more difficult (see below). Using statistics collected from tens-of-thousands of coincidence events, a set of simultaneous equations for the total activity of each parcel of tissue along many LORs can be solved by a number of techniques, and thus a map of radioactivities as a function of location for parcels or bits of tissue ("voxels"), may be constructed and plotted. The resulting map shows the tissues in which the molecular probe has become concentrated, and can be interpreted by nuclear medicine physician or radiologist in the context of the patient's diagnosis and treatment plan.

Imaging

Preprocessing Filtered back projection (FBP) has been frequently used (simple but sensitive to low level discrete noise (shot noise) forming streaks across the image). Iterative expectation-maximization algorithms are the preferred method of reconstruction but requires higher computer power. (For these techniques see imaging textbooks and for instance Wikipedia.)

Attenuation correction As different LORs must traverse different thicknesses of tissue, the photons are attenuated differentially. The result is that structures deep in the body are reconstructed as having falsely low tracer uptake. Contemporary scanners can estimate attenuation using integrated CT equipment. Since correction is itself susceptible to significant artifacts both corrected and uncorrected images are always read together.

2D/3D reconstruction Modern scanners have a cylinder of scanning rings. There are two approaches: individual reconstruction per ring (2D reconstruction) or allow coincidences to be detected between rings as well as within rings, then reconstruct the entire volume together (3D). 3D techniques have better sensitivity (because more coincidences are detected and used) and therefore less noise, but are more sensitive to the effects of scatter and random coincidences, as well as requiring correspondingly greater computer resources.