How is data acquired in SPECT?
In SPECT imaging, data are acquired by rotating one or more gamma camera heads around the patient and recording projection images at multiple angular positions. Each projection represents the spatial distribution of detected gamma photons from a specific angle. These projections are then reconstructed into cross-sectional slices, allowing three-dimensional visualisation of tracer distribution.
SPECT acquisition involves rotating the gamma camera to collect multiple angular projections, with image quality determined by angular sampling, count statistics, and detector geometry.
The camera typically rotates through 180° or 360°, acquiring images at regular angular intervals. At each position, counts are collected for a fixed duration before the camera moves to the next angle. The total number of projections and the time spent at each position directly influence image quality.
Understanding the physics
SPECT acquisition builds on planar gamma camera imaging. In planar imaging, a single projection represents the summed activity along each line through the body. In SPECT, multiple such projections are collected as the detector rotates around the patient. Each projection captures information from a different viewing angle, and the combination of these projections allows reconstruction of the underlying three-dimensional activity distribution.
The number of angular samples is critical. If too few projections are acquired, reconstruction artefacts may appear because the system lacks sufficient information to accurately localise activity. Increasing the number of projections improves sampling but increases total scan time.
At each angular position, counts are accumulated for a defined period. Longer acquisition at each angle increases total detected counts and improves signal-to-noise ratio, consistent with Poisson statistics. However, extending acquisition time increases the risk of patient motion, which can degrade reconstructed images.
Matrix size also influences acquisition characteristics. A larger matrix improves spatial sampling but distributes counts over more pixels, increasing noise per pixel unless total counts are increased accordingly. Smaller matrices improve count density but reduce spatial sampling precision.
Detector rotation may follow a circular orbit or adjust dynamically to remain as close as possible to the patient. Reducing detector-to-patient distance improves spatial resolution because collimator resolution degrades with increasing distance.
SPECT acquisition represents a balance between angular sampling, count statistics, spatial resolution, and scan duration.
Where this matters clinically
Optimising SPECT acquisition parameters is essential for high-quality myocardial perfusion, brain perfusion, and skeletal studies. Insufficient counts or inadequate angular sampling can produce noisy or streaked reconstructions, while excessive acquisition times increase motion artefact risk.