Syngo trued

The MR and PET images were co-registered using Syngo.via and Syngo TrueD systems (Siemens Healthcare, Malvern, PA) software, which was then used for the measurement of the average standardized uptake value (SUV) in a 1 cm³ spherical region of interest (ROI) around the visually determined lesions and in the anatomically equivalent contralateral brain region. The SUV represents the radioactivity concentration observed in an ROI relative to the hypothetical case of homogeneous distribution of the injected radioactivity across the entire body of the patient. In this study, SUV_DVA is defined as the mean SUV from the region of the DVA divided by the mean SUV in the anatomically equivalent contralateral region. The size of each DVA was measured as the single longest dimension of the aberrant vasculature. The qualitative characterization of the degree of metabolic abnormality (mild, moderate, or severe hypometabolism; isometabolism; or mild, moderate, or severe hypermetabolism) was determined by a consensus of the authors.

An experienced nuclear medicine specialist, who also identified all PCa lesions in the pelvis, qualitatively evaluated the reconstructed PET images. For a quantitative assessment, we analyzed the SUV_mean and SUV_max for each detected lesion located in the entire pelvis, as well as the SUV_mean of its respective background (BG), using the PERCIST reference volumes of interest with commercial software (syngo TrueD, Siemens Healthcare, Erlangen, Germany).
In the absence of absolute ground truth data, the results were compared using linear regression against standard Dixon-VIBE, which was available on the system from the very beginning. Additionally, the Wilcoxon signed-rank tests were performed to test whether the SUV_mean and SUV_max values in PET images reconstructed using each AC method were significantly different from the corresponding values in the PET images reconstructed with other evaluated methods. We investigated variance and correlation between the methods and calculated the coefficient of correlation r as well as the coefficient of determination R² (“R squared”) for each evaluated method against standard Dixon-VIBE. The differences between the AC methods, i.e., the estimated bias and fluctuations in SUV_max, were additionally visualized and evaluated using the Bland-Altman plots (the agreement limits were defined by the 96% confidence level).

Quantitative measurements were performed on a dedicated workstation (Syngo TrueD, Siemens Medical Solutions).
Volume of interest (VOI) was performed using a 40% threshold of SUVmax. The SUVmax, SUVmean and MTV were obtained for each identified lung lesion. TLG defined as MTV multi-plied by the average SUV uptake (SUVmean) within the MTV. We used the liver as a source for the background and noise measurement [13 (link)]. In each patient, a 30 mm-diameter spherical VOI was placed within an area of uniform FDG distribution in the liver, the mean SUV and standard deviation within the VOI were obtained. The signal-to-nose ratio of the tumour, relative to the liver, SNR (T–L) was calculated as: $$SN{R}_{(T-L)}=\frac{Tumour-Liver}{S{D}_L}$$ where the Tumour refers to SUVmax in the lung lesion, Liver and SD_L is the mean SUV and standard deviation measured in the liver VOI respectively.

PET/CT Images will be reviewed and analyzed using Siemens Syngo/TrueD and OSIRIX workstations by a board certified nuclear medicine physician and a board certified radiologist experienced in reading PET/CT using recent reporting guidelines (PROMISE criteria, miTNM standardized framework) [50 (link)].