Abstract
Our purpose is to discuss two types of corrections for count loss that occur while performing SPECT studies. These corrections will improve image quality. Radioactive decay correction should be considered with the development of new short-lived radiopharmaceuticals for SPECT imaging, such as technetium-99m (99mTc) HMPAO. The count rate is reduced by approximately 8.3% from radioactive decay of 99mTc during a standard 45-min SPECT acquisition. This can easily be corrected by multiplying each frame in the raw data by the decay factor. Count loss also occurs due to the attenuation of the pallet as the camera rotates under the patient. Pallet attenuation can be visualized and calculated by doing a 360° acquisition of a SPECT phantom filled with 99mTc. We have evaluated the attenuation from different manufacturer’s pallets and head holders. There is as much as a 12.6% loss in count rate from the Siemens heavy-duty pallet and as little as 3.5% from the G.E. attachable head holder. A computer algorithm could be developed to correct for pallet attenuation, but the problem can be avoided by changing the system design for different types of SPECT acquisitions, especially when scanning the brain or heart. Some departments begin their SPECT acquisition in the posterior or lateral position and thereby compound the decay and pallet attenuation problems by causing the last few frames of the study to be acquired directly beneath the pallet where maximum attenuation will occur during maximum decay loss. Both the decay and pallet attenuation problems can combine to create as much as a 15%-20% loss of counts from the first to last image in a given 99mTc SPECT study.
