Detectors

To image the distribution of positron-emitting isotope in the body, both of the 511 keV photons emitted from positron annihilation must be detected in coincidence. Unlike other instruments used in nuclear medicine, PET uses electronic rather than lead collimators to detect signal (event) results from annihilation of the positron and an electron. The probability of detecting both photons depends on the detector efficiency, which is strongly related to the stopping power of the scintillator and the thickness of the scintillator used in the detector. Early generation of PET scanners used NaI(Tl) crystals, the same material used in gamma camera. Modern PET scanners use much denser scintillators, such as bismuth germanate oxide (BGO) [27], which has been the scintillator of choice for more than two decades due to its very high density and stopping power for the 511 keV gamma rays. In order to provide higher detection efficiency and spatial resolution with lower production cost, a number of detector designs were proposed in the 1980s and the most successful one was the block detector technique proposed by Casey and Nutt, using BGO crystal [28]. A typical BGO block detector comprises a rectangular block consisting of between 6 x 8 and 8 x 8 individual scintillation crystals, attached to an array (usually 2 x 2) of photomultiplier tubes (PMTs) at which the scintillation light is amplified and converted into electrical signal for the coincidence detection circuit to register. A schematic outline of such a block detector is shown in Fig. 2.3. The BGO element in which a gamma ray interacts is determined by the relative light output

Scintillator array

PMTs

PMTs

Figure 2.3: Schematic diagram of a BGO block detector commonly used in commercial PET systems.

from the four PMTs. Anger-logic is used to obtain the X and Y positions based on the four PMT outputs Pi:

The combined BGO block/photomultiplier system has an approximately cubic spatial resolution of 4 mm full-width-at-half-maximum and coincidence timing window of approximately 12 ns.

As seen from Fig. 2.2, the probability that the annihilation event occurs exactly within the region of coincidence detection and is recorded by the detectors is very small because most gamma rays may travel out of the region of coincidence detection even if the annihilation event occurs within that region. This probability can be increased by using a ring of detectors within which any detector is in coincidence with all other detectors located at the opposite side of the ring. With the use of multiple rings of detectors, the probability of coincidence detection is further increased because coincidences can be detected by other rings of detectors if they cannot be recorded by the plane of the ring within which the annihilation events occur. The device that used to detect the 511 keV gamma rays emitted from annihilation and construct a map of radiopharma-ceutical distribution inside the body is called tomograph (or scanner), which usually has multiple rings of detectors surrounding the patient.

Was this article helpful?

0 0

Post a comment