Count-rate analysis from clinical scans in PET with LSO detectors

[Optimization of the scan time is based on the physical index in FDG-PET/CT]

PURPOSE

The image quality of Positron Emission Tomography (PET) using (18)F-fluoro-2-deoxy-D-glucose is extremely important to diagnose cancer precisely. The purpose of this study is to inspect the physical indexes that are correlated with image quality and optimization of the scan time, utilizing the physical index.

METHODS

This study calculated two kinds of patient noise equivalent counts (NEC(patient) and NEC(density)) and signal to noise ratio in liver (SNR(liver)) to measure PET image quality. We estimated the correlation coefficient between the visual assessment of PET image quality and physical indexes to assess the degree of correlation. We also set the optimal scan time, depending on body mass index (BMI), using the physical index and examined the efficacy of the optimization.

RESULTS

NEC(density) showed the highest correlation coefficient in PET image quality (r=0.743, p<0.001). By optimizing the scan time using NEC(density), we showed there was no correlation between BMI and NEC(density) (r=0.192, p=0.047). In addition, reduced dependence of the PET image quality by BMI was showed.

CONCLUSION

The optimization of the scan time using NEC(density) reduces deterioration of the PET image quality by patient habitus.

Comparison of imaging protocols for 18F-FDG PET/CT in overweight patients: optimizing scan duration versus administered dose

The quality of (18)F-FDG PET/CT images of overweight patients is often degraded. We evaluated the effect of optimizing injected dose or acquisition time on the quality of images of overweight patients using lutetium oxyorthosilicate PET/CT with high-performance detector electronics.

METHODS

We initially retrospectively measured radioactivity concentrations and signal-to-noise ratios (SNRs) in the liver relative to body weight for 80 patients who had undergone (18)F-FDG PET/CT according to our standard protocol (injected dose, 3.7 MBq/kg; acquisition time, 2 min/bed position). The patients were grouped (n = 20 per group) according to baseline body weight as G1 (<or=59 kg), G2 (60-69 kg), G3 (70-84 kg), and G4 (>or=85 kg). We compared the SNRs of G1 with those of G2, G3, and G4 and calculated the ratio squared as a factor to correct the acquisition parameters for overweight patients. We then prospectively enrolled 120 patients according to the same body weight criteria. We multiplied the correction factors to optimize injected doses or acquisition times and defined dose-adjusted groups (n = 20 per group) and time-adjusted groups (n = 20 per group). G2 dose was defined as 5.59 +/- 0.19 MBq/kg, G3 dose as 7.29 +/- 0.33 MBq/kg, and G4 dose as 8.88 +/- 0.43 MBq/kg. G2 time was defined as 3 min/bed position, G3 time as 4 min/bed position, and G4 time as 5 min/bed position.

RESULTS

Although liver activities did not significantly differ among G1 through G4 irrespective of patient weight, SNR progressively decreased as patient weight increased. The liver activities of G2 dose, G3 dose, and G4 dose were, respectively, 1.4-, 1.9-, and 2.5-fold higher than those of the baseline counterparts. Nevertheless, the increased liver activities of G2 dose, G3 dose, and G4 dose did not significantly affect SNR, compared with the baseline groups. In contrast, the SNR of G4 time was significantly higher than that of G4.

CONCLUSION

Our findings suggest that the quality of images acquired from heavier patients can be maintained only by scanning for longer periods. Increasing the dose per kilogram of body weight did not improve the quality of lutetium oxyorthosilicate PET/CT images.