Scan Provides High-Quality Images of Small Tumors
Combining technologies improves SPECT image quality, say researchers at SNM’s 56th Annual Meeting
A new study shows that combining high resolution and high sensitivity collimation provides better quality images when using single photon emission computed tomography (SPECT) scans, said researchers at SNM’s 56th Annual Meeting. Collimators””devices that filter a stream of rays so that only those traveling parallel to a specified direction are allowed through””are used in SPECT because it is not yet possible to focus radiation with such short wavelengths into an image with the use of lenses. Using the two types of collimation in tandem is especially effective for better imaging of small tumors, the study indicates.
"SPECT is an important tool in molecular imaging because of its ability to provide accurate images of what is going on in the body without the need for invasive procedures such as surgery," said Roel Van Holen, researcher in the department of electronics and information systems of Ghent University, Ghent, Belgium, and lead author of the study. "However, researchers have had to make tradeoffs in SPECT image quality, especially in imaging very small tumors. Our research is exciting because it shows that combining high-resolution and high-energy collimators can improve SPECT’s ability to image small tumors."
Researchers have established that SPECT image quality generally improves when using collimators with higher sensitivity than traditional low-energy high-resolution (LEHR) collimators. However, this is not the case when small tumors are being imaged. In these cases, LEHR actually provides better images.
The new study combined a high-resolution collimator with a high-sensitivity collimator. A dual-head SPECT camera with three different collimator settings was simulated using the GATE Monte Carlo simulator. The results indicated that SPECT image quality was better with the mixed collimation than it was by using only high-resolution or only high-sensitivity collimation.
On The Net: