The aim of this study was to investigate the impact of time-of-flight (TOF) on quantification and reduction of respiratory artifacts.
The National Electrical Manufacturers Association phantom was used for optimization of reconstruction parameters. Twenty seven patients with lesions located in the diaphragmatic region were evaluated. The PET images were retrospectively reconstructed using non-TOF (routine protocol in our department) and TOF algorithms with different reconstruction parameters. Maximum standardized uptake value, estimated maximum tumor diameter, coefficient of variation, signal-to-noise ratio, and lesion-to-background-ratio were also evaluated.
On the basis of phantom experiments, TOF algorithms with two iterations, 18 subsets, and 5.4 mm and 6.4 mm postsmoothing filter reduced the noise by 3.1 and 12.6% in phantom with 2 : 1 activity ratio, and 3.0 and 13.1% in phantom with 4 : 1 activity ratio. The TOF algorithm with two iterations, 18 subsets, and 6.4 mm postsmoothing filter had the highest signal-to-noise value, and was selected as the optimal TOF reconstruction. Mean relative difference for signal-to-noise between non-TOF and optimal TOF in phantom with 2 : 1 and 4 : 1 activity ratio were 11.6 and 18.7%, respectively. In clinical data, the mean relative difference for estimated maximum tumor diameter and maximum standardized uptake value between routine protocol and optimal TOF algorithm were −6.3% (range: −20.4 to −0.6%) and 13.2% (range: 0.3–57.6%), respectively.
Integration of TOF in reconstruction algorithm remarkably improved the white band artifact in the diaphragmatic region. This technique affected the quantification accuracy and resulted in smaller tumor size and higher standardized uptake value in tumors located in/near the diaphragmatic region.
aDepartment of Medical Physics and Biomedical Engineering
bResearch Center for Molecular and Cellular Imaging, Tehran University of Medical Sciences
cChronic Respiratory Diseases Research Center, National Research Institute of Tuberculosis and Lung Diseases (NRITLD), Shahid Beheshti University of Medical Sciences
dPET/CT and Cyclotron Center, Masih Daneshvari Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran
Departments of eRadiology
fElectrical and Computer Engineering, Johns Hopkins University, Baltimore, Maryland, USA
Correspondence to Pardis Ghafarian, PhD, Chronic Respiratory Diseases Research Center, National Research Institute of Tuberculosis and Lung Diseases (NRITLD), Shahid Beheshti University of Medical Sciences, Tehran 19569-44413, Iran Tel/fax: +98 912 504 7908/+98 212 610 9484; e-mail: email@example.com
Received March 9, 2017
Received in revised form August 5, 2017
Accepted August 7, 2017