Optimizing positron emission tomography image acquisition protocols in integrated positron emission tomography/magnetic resonance imaging

Abstract

Objectives: In integrated positron emission tomography (PET)/magnetic resonance imaging (MRI), the PET data acquisition is performed simultaneously to the magnetic resonance data acquisition, leaving latitude for the duration of PET acquisition time. This establishes emission time as an important parameter in forthcoming PET/MRI protocols because it is one of the key factors determining PET image quality. Thus, the purpose of the current study was to identify optimal duration of PET acquisition time in PET/MRI.

Materials and methods: A total of 22 consecutive patients (7 men, 15 women) underwent fluorine-18-labeled fluorodeoxyglucose (18[F]FDG) PET/MRI after clinical PET/computed tomography. Positron emission tomography/magnetic resonance scans were acquired for 8 minutes per bed position (mpb). Positron emission tomography was extracted to reconstruct images with 2, 4, 6, and 8 mpb for each patient. Visual and quantitative approaches were used to assess image quality and lesion detectability for each image. For image quality, (a) 3 readers independently scored subjective image quality on a 4-point scale and (b) a region-of-interest approach was used to obtain a quantitative estimate of image quality in terms of noise. For lesion detectability, (a) the readers independently counted the number of hypermetabolic lesions and (b) signal-to-noise ratio and contrast-to-noise ratio were computed in a region-of-interest approach. Moreover, the mean and maximal standardized uptake value (SUV mean and SUV max, respectively) of the hypermetabolic lesions was compared across all acquisition times.

Results: For image quality, subjective image quality significantly declined from 8 to 2 mpb (P <; 0.05), with the exception of the difference between 6 and 8 mpb. Image noise increased with shorter imaging duration, ranging from 13% on average in the 8-mpb scans to 23% in the 2-mpb scans (differences were statistically significant for 2 vs 6 mpb, 2 vs 8 mpb, and 4 vs 8 mpb; P <; 0.05). For lesion detectability, 39 hypermetabolic lesions were identified by consensus. There was no difference in detected lesions across all acquisition times. Signal-to-noise ratio and contrast-to-noise ratio were constantly high and did not differ significantly across the acquisition times. The SUV mean and SUV max did not differ significantly across all acquisition times.

Conclusions: Positron emission tomography acquisition times on integrated PET/MRI do not need to exceed usual acquisition times on current PET/computed tomography scanners: Although the PET image quality suffers from short acquisition times, even a duration of 2 mpb permits sufficient lesion detection. Moreover, quantitative measures of tracer uptake are also reasonably precise at short acquisition times.