Abstract
Kinetic modeling of PET-18F-FDG in the arteries has to be approached in a different manner than for any other organ like the brain, the liver or for a tumor. The artery PET images contain more activity signals coming from blood circulating throughout the lumen and activity from the surrounded thin layer of tissue that is subjected to partial volume effect (PVE).
In the present work, eleven subjects were measured with PET/CT and 18F-FDG in dynamic mode. With CT imaging, the arteries with calcifications were evidenced, and glucose uptake was expected to be dependent on calcification intensity in Hounsfield units (HU) and on calcification volume (area ratio of calcification), which are both related to the degree of artery inflammation. 18F-FDG uptake was assessed with the fractional uptake rate (FUR) and the calcification intensity on CT images was classified in four levels. PET artery images were corrected for PVE based on their corresponding size in CT images. Furthermore, 18F-FDG images were decomposed with Factor Analysis (FA) in blood and tissue components. The results showed that the total analyzed artery segments were 1015, 708 of them were not calcified (69.8%) and 307 (30.2%) were found with calcification. FUR values were statistically significant between non-calcified and calcified artery segments (p < 0.05; mean non-calcified = 0.0827; mean calcified = 0.0727). The FUR calculation also showed significant differences at the higher calcification areas. In conclusion, it is possible to depict the differences in artery uptake of 18F-FDG when correcting for PVE, extracting the input function with FA and applying kinetic modeling on dynamic images.