40.3.1 Types of coincidence

For the annihilation radiation from the positron–electron interactions to give useful information, the two photons must travel in straight lines from the source of the radiation to the detectors. The photons in (A) of Figure 40.2 are of this type.

Figure 40.2 Types of coincidence. (A) is a true coincidence. In (B), one of the photons is scattered and in (C), one of the photons is absorbed. This makes (B) and (C) false coincidences.

Because the photons all travel with the velocity of electromagnetic radiation, they will each be detected by a detector almost at the same time – within <10×10⁻⁹ sec of each other. These photons are said to be coincident. It is possible for one of the photons to be scattered (see Fig. 40.2B). If the position of these two photons were back- projected, it would be an incorrect position for the origin of the annihilation radiation. Because the scattered photon travels along a slightly longer route, the two photons will not be detected within 10×10⁻⁹ sec of each other and so they are not said to be coincident and the imaging computer can be programmed to ignore them. Similarly, in Figure 40.2C, one of the photons has been absorbed. Because there is no matching coincident photon, the imaging computer will again ignore this event.

40.3.2 Detector materials

In the gamma camera (see Ch. 37), the radiation is detected using a sodium iodide crystal. The radiation from technetium-99m has an energy of 140 keV. The energy of the annihilation radiation from the positron–electron interaction is 511 keV. The greater energy of this radiation requires a material of higher density and/or higher atomic number. Bismuth germinate (BGO) is the material of choice for the detector crystals.

40.3.3 Detector mechanisms

The basic components of the detector mechanism are BGO crystals over which photomultiplier tubes or photodiodes are positioned as shown in Figure 40.3. These arrangements are known as blocks. Each photomultiplier tube detects the light emission as the result of the radiation photons interacting with the crystal. Groups of blocks (see Fig. 40.4) arranged with shared electronics form detector cassettes and a number of these detector cassettes form a detector ring around the patient.

Figure 40.3 Arrangement of BGO crystal and photomultiplier tubes to make an imaging block.

Figure 40.4 Arrangement of imaging blocks to make an imaging ‘cassette’.

Normally there is more than one ring so that we can detect information from a number of slices simultaneously. To avoid oblique rays from one slice interfering with the image on another ring, the rings are separated by septa made of lead or tungsten (see Fig. 40.5).

Figure 40.5 Arrangement of detection rings separated by septa which avoid detection of oblique rays.

40.5.1 Time of flight (TOF) profiles

If we consider the method of producing images of the positron annihilation discussed so far, the radiation is detected by two detectors thus establishing the line in which the positron emission has taken place. The image of this event is then produced by ‘back projection’ of the information along the whole of this line. In TOF scanners, the precise time taken for each of the photons to reach its detector is measured. As can be seen in Figure 40.6 (see page 300), the distance D₁ is greater than the distance D₂ so this first photon will take slightly longer to reach the detector than the second photon. If we know the velocity of electromagnetic radiation (3×10⁸ m.s⁻¹) then we can get a more accurate calculation of the distance travelled by each photon. Thus, in the back projection of the image, instead of simply saying the photons were emitted somewhere along this line, we can say more precisely the position of the emitted radiation on the line. This is illustrated in Figure 40.7 (see page 300).

Figure 40.6 Line diagram to explain TOF technology. The distances D₁ and D₂ from the source of activity to the detectors is shown by the open arrows.

Figure 40.7 Back-projected image showing the probable position of the area of increased activity from the TOF calculations.

No timer is 100% accurate so there is always some degree of positional uncertainty. The fact that we are able to more accurately calculate the positions of the emitted radiation photons means that we get improved resolution in the image.

40.5.2 Hybrid scanners

PET scanning gives us information regarding the position of high physiological activity. It, however, gives little or no information about related normal anatomy. This makes the interpretation of the images and the management of the patient a more difficult task. For this reason, scanners have been constructed which give anatomical information (e.g. from a computed tomography or magnetic resonance imaging scan) and physiological information (e.g. from a PET or single photon emission computed tomography (SPECT) scan). The images from each modality may be superimposed to give a composite image of the anatomy and physiology of a region. These scanners are known as hybrid scanners and will form the topic for Chapter 41.