Assessment of reduced 18F-FDG dosage in pediatric PET/MR imaging through randomized down-sampling.

Abstract

Purpose: This study aimed to explore the feasibility of reduced ¹⁸F-FDG dosage in pediatric PET/MR imaging through randomized down-sampling.

Methods: Forty-four pediatric patients (26 males, 18 females; median: 11 years [range, 1-17]) underwent ¹⁸F-FDG PET/MR examinations were retrospectively enrolled. PET data were reconstructed into 6 groups with different simulated doses using randomly down-sampling: full dose (2.96 MBq/kg), 1/2 dose, 1/3 dose, 1/4 dose, 1/10 dose, and 1/20 dose. Subjective visual analysis of image quality was evaluated using a 5-point Likert scale. Quantitative parameters of lesions and blood pool and liver background were analyzed. The full dose group was served as a reference.

Results: Subjective image quality of 1/2 dose, 1/3 dose and 1/4 dose images was clinically acceptable (score ≥ 3), and all FDG-avid lesions could be identified in these three groups like full-dose group. SUVmax and standard deviation (SD) of blood pool and liver increased as the dose reduced, while no significant difference in SUVmean of blood pool and liver was found in 1/2-1/20 dose groups when compared to the full dose group. Besides, a significant difference was observed in SUVmax and SD of lesion and lesion-to-background ratio between 1/2-1/20 dose and full dose groups (p < 0.05).

Conclusion: Further reduction in administered ¹⁸F-FDG activities is feasible in pediatric PET/MR scan. Clinically acceptable image quality could be achieved using a low dose of 0.74 MBq/kg.