TABLE OF CONTENTS Ideal PET imaging requires that neither of the annihilation photons undergo any interaction in the biological tissue. By neglecting the photon absorption within the FOV, quantitative measurements are not possible: the number of counts does not correspond to the number of emitted positrons. In PET, an attenuation correction factor (ACF) should be used for each LOR: it is theoretically equal to the inverse of the attenuation probability along the LOR.
The ACF is easily measured for a phantom that presents a uniform linear attenuation coefficient ?, by using the exponential attenuation law. The first photon covers the distance d 1 and its probability to hit the detector without any interaction in the object is proportional to e ?? d 1; the second one covers the distance d 2 with an associated probability proportional to e ?? d 2. Thus, the probability that both the photons are detected without any interaction is proportional to e ??( d 2+ d 2) = e ?? d. The ACF is e ? d, which is independent from the position of the annihilation point, and is a function only of the thickness d of the phantom along the LOR. Thus, the correction needs only a sampling of the object thickness in the FOV (Fig. 8.30). This fast correction is used in examinations where the ? value is only approximately uniform, as in human head and...
