Technical Principle of PET (Positron emission tomography)

 Positron emission tomography

 Positron emission tomography is an imaging modality for obtaining in vivo cross-sectional images of positron-emitting isotopes that demonstrate biological function, physiology, or pathology.
 In this technique, a chemical compound with the desired biological activity is labeled with a radioactive isotope that decays by emitting a positron or positive electrons.
 The emitted positron almost immediately combines with an electron and the two are mutually annihilated with the emission of two gamma rays.
 The two gamma-ray photons travel in almost opposite directions, penetrate the surrounding
tissue, and are recorded outside the subject by a circular array of detectors.

 A mathematical algorithm applied by the computer rapidly reconstructs the spatial distribution of the radioactivity within the subject for a selected plane and displays the resulting image on the monitor.

 Thus, PET provides a non-invasive regional assessment of many biochemical processes that

are essential to the functioning of the organ being visualized.

 The positron(b+ ) is emitted from a proton-rich nucleus with a variable amount of kinetic Energy.

 This energy is dissipated in the patient over a range of tissue of the order of a few millimeters.

 The b+ combines with a free electron (b– ) and the masses are transmuted to two 511-keV g rays

which are emitted at 180° ± 0.25° to one another to satisfy the conservation of momentum as in Fig.21.18.

 The finite variable range of the b+ as well as the angular variation of about 180o is a fundamental

limitation to the resolution achievable with PET.

 The compounds used and quantitated are labeled with proton-rich positron (b+ ) emitters that are usually cyclotron-produced.

 A variation of 0.25° in the angular distribution of the back-to-back 511 keV g rays will cause a degradation of 1.75 mm at the center of an 85-cm whole-body tomograph.

 Neither of the primary limitations will cause a significant loss in resolution in present-day PET designs.

 They are simply fundamental to the method and cannot be eliminated.

 Two design types of positron-emission tomographs have been introduced, one employing opposed large-area detectors which require rotation around the patient to provide the necessary degree of angular sampling, and the other, employing multiple individual crystal detectors surrounding the patient in a circular or hexagonal array.

 Conventional lead absorption collimators are not required because the coincident detection of two

511 keV photons indicate the line of origin along which the photons were emitted.

 However, in order to reduce the random coincidence count rate, some degree of collimation is normally employed.

 Pulse processing needs to be much faster than with single-photon systems, to keep random coincidences to manageable proportions.

 With fast-response detectors and suitably fast electronics, it may be possible to use the difference in the time of arrival of the annihilation photons at opposite detectors to locate the site of positron decay and improve spatial resolution.

Figure 21.19 illustrates the gantry and detector components used by Hoffman et al (1985) in a PET system.

 The gantry has a large opening (diameter = 65 cm) and can image both the brain and torso of an adult patient.

 The entire detector assembly may be tilted to obtain oblique sections. Bismuth germanate (BGO) scintillation crystals, 5.6 mm wide, 30 mm high, and 30 mm deep, are used to detect the 511 keV annihilation radiation.

The detectors are arranged in a circular ring geometry, with 512 detectors per ring. The system has two rings and produces three scanning planes (two direct and one cross-plane).

 In order to facilitate replacement, the detectors are arranged in modules or buckets containing 16 detector packages.

 Each package contains two crystals and two PMTs.

 The center-to-center spacing of the crystals is 6.1 mm. Axially, the two rings are separated by 36 mm.

 Besides containing the two BGO crystals and PMTs, the bucket also contains amplifiers/ discriminators and other front-end processing electronics.

 In order to increase linear sampling, the entire detector assembly can wobble in a small circular orbit.

 This wobbling procedure is used to optimize spatial resolution (Jaszczak, 1988).

 A simplified block diagram of the data acquisition system is shown in Fig. 21.20. Distributed processors are used throughout the system to maximize speed for simultaneous data collection.

 Individual and analog detector signals are amplified and the time of interaction is then determined using a constant fraction discriminator.

 A time encoder converts the event into a 14-bit word containing the detector number and event time within 8 ns.

 This word is passed to the coincidence processor

every 224 ns.

 The energy window is controlled automatically by the microprocessor located in each detector bucket.

 A threshold of 200 keV is typically used to allow for the detection of gamma rays that have been scattered within a detector and escaped.

 The system consists of a fan beam geometry with an angular sampling of 0.7 degrees.

 The linear sampling is 2.9 mm. The main processor serves to monitor and control the various processing jobs.

 An array processor is used to perform the primary reconstruction.

 Several peripheral devices, including a display processor, are attached to the system computer.