The principle of PET imaging

Primarily, let us consider the basic principle of PET measurements: the concentration of the marker is higher at the critical spots (e.g. in the case of FDG (fluorodeoxyglucose) because of the increased sugar consumption of the cancerous cells), and when the positron created during the decay of the isotope in the marker meets an electron, they annihilate, resulting in the simultaneous release of two 511 keV gamma photons travelling in nearly opposite directions. If we detect these pairs, we can determine the activity concentration using the density of the LORs (line of response) connecting the detection points. The basic principle of measurement is depicted in the figure below.

Figure 1.: The measured phenomenon, the measured data and the determinant activity concentration. This figure was drawn based on the lecture of Sibylle Ziegler

Of course, the efficiency of the detectors is not 100%, they do not detect every gamma photon: as the penetrating power of high-energy photons is relatively high, in some cases they pass through the detector without interaction. It is possible that by detecting two photons of two different ‘half-detected’ photon pairs simultaneously, we might believe that the two photons originate from a single decay (this is called accidental coincidence). Alternatively, they may get absorbed or scattered in the medium we want to examine, thus their direction changes. Scattering is especially important in human PET scans, since the size of the human body scatters with a much higher probability than e.g. a mouse. The accidental coincidence and scattering can cause the problems depicted in Figure 2.

Figure 2.: False LORs caused by random and scattered events. The point of the decay is indicated by red dots, the gamma photon pairs are indicated by red arrows and the apparent LORs are indicated by green arrows.