The imaging process of SPECT can be simply depicted as in Figure 13.6. Gamma-ray photons emitted from the internal distributed radiopharmaceutical penetrate through the patient’s body and are detected by a single or a set of collimated radiation detectors. The emitted photons experience interactions with the intervening tissues through basic interactions of radiation with matter [Evans, 1955]. The photoelectric effect absorbs all the energy of the photons and stops their emergence from the patient’s body. The other major interaction is Compton interaction, which transfers part of the photon energy to free electrons. The original photon is scattered into a new direction with reduced energy that is dependent on the scatter angle. Photons that escape from the patient’s body include those that have not experienced any interactions and those which have experienced Compton scattering. For the primary photons from the commonly used radionuclides in SPECT, for example, 140-keV of TC-99m and ∼70-keV of TI-201, the probability of pair production is zero.
Most of the radiation detectors used in current SPECT systems are based on a single or multiple NaI(TI) scintillation detectors. The most significant development in nuclear medicine is the scintillation camera (or Anger camera) that is based on a large-area (typically 40 cm in diameter) NaI(TI) crystal [Anger, 1958, 1964]. An array of photomultiplier tubes (PMTs) is placed at the back of the scintillation crystal. When a photon hits and interacts with the crystal, the scintillation generated will be detected by the array of PMTs. An electronic circuitry evaluates the relative signals from the PMTs and determines the location of interaction of the incident photon in the scintillation crystal. In addition, the scintillation cameras have built-in energy discrimination electronic circuitry with finite energy resolution that provides selection of the photons that have not been scattered or been scattered within a small scattered angle. The scintillation cameras are commonly used in commercial SPECT systems.
Analogous to the lens in an optical imaging system, a scintillation camera system consists of a collimator placed in front of the NaI(TI) crystal for the imaging purpose. The commonly used collimator is made of a large number of parallel holes separated by lead septa [Anger, 1964; Keller, 1968; Tsui, 1988]. The geometric dimensions, that is, length, size, and shape of the collimator apertures, determine the directions of photons that will be detected by the scintillation crystals or the geometric response of the collimator.
The width of the geometric response function increases (or the spatial resolution worsens) as the source distance from the collimator increases. Photons that do not pass through the collimator holes properly will be intercepted and absorbed by the lead septal walls of the collimator. In general, the detection efficiency is approximately proportional to the square of the width of the geometric response function of the collimator. This tradeoff between detection efficiency and spatial resolution is a fundamental property of a typical SPECT system using conventional collimators.
The amount of radioactivity that is used in SPECT is restricted by the allowable radiation dose to the patient. Combined with photon attenuation within the patient, the practical limit on imaging time, and the tradeoff between detection efficiency and spatial resolution of the collimator, the number of photons that are collected by a SPECT system is limited. These limitations resulted in SPECT images with relatively poor spatial resolution and high statistical noise fluctuations as compared with other medical imaging modalities. For example, currently a typical brain SPECT image has a total of about 500K counts per image slice and a spatial resolution in the order of approximately 8 mm. A typical myocardial SPECT study using TI-201 has about 150K total count per image slice and a spatial resolution of approximately 15 mm.
In SPECT, projection data are acquired from different views around the patient. Similar to x-ray CT, image processing and reconstruction methods are used to obtain transaxial or cross-sectional images from the multiple projection data. These methods consist of preprocessing and calibration procedures before further processing, mathematical algorithms for reconstruction from projections, and compensation methods for image degradation due to photon attenuation, scatter, and detector response.
The biokinetics of the radiopharmaceutical used, anatomy of the patient, instrumentation for data acquisition, preprocessing methods, image reconstruction techniques, and compensation methods have important effects on the quality and quantitative accuracy of the final SPECT images. A full understanding of SPECT cannot be accomplished without clear understanding of these factors. The biokinetics of radiopharmaceuticals and conventional radiation detectors have been described in the previous section on conventional nuclear medicine. The following subsections will present the major physical factors that affect SPECT and a summary review of the instrumentation, image reconstruction techniques, and compensation methods that are important technological and engineering aspects in the practice of SPECT
