%0 Journal Article %A Lucia Baratto %A Heying Duan %A Harsh Gandhi %A Praveen Gulaka %A Andrei Iagaru %T The effect of various beta values on image quality and semi-quantitative measurements in 68Ga-labeled GRPR and PSMA PET/MRI images reconstructed with a block sequential regularized expectation maximization algorithm %D 2018 %J Journal of Nuclear Medicine %P 1751-1751 %V 59 %N supplement 1 %X 1751Objectives: The block sequential regularized expectation maximization (BSREM) algorithm is a new image reconstruction method that controls noise at higher iterations by applying a relative difference penalty built into the objective function. This enables one to employ more iterations for convergence and better contrast recovery, while mitigating noise amplification. BSREM was recently introduced on the GE SIGNA PET/MRI platform. Here we evaluated how different β values influence image quality and maximum standardized uptake values (SUVmax) in a cohort of prostate cancer patients who underwent 68Ga-RM2 or 68Ga-PSMA-11 scans. Materials and methods: We analyzed data from 36 participants with prostate cancer who underwent either 68Ga-RM2 (15) or 68Ga-PSMA-11 (21) PET/MRI. The raw PET data were retrospectively reconstructed using β values of 250, 350, 500, 750 and 1000. Each reconstruction was reviewed blindly, independently and in random order by 3 nuclear medicine physicians and scored using a Likert scale (1 - poor, 5 - excellent quality). Maximum standardized uptake values (SUVmax) were measured from 68Ga-RM2 PET/MRI and 68Ga-PSMA PET/MRI for all the lesions identified as compatible with prostate cancer. Results: The mean±SD scores for 68Ga-RM2 PET images were 2.5±0.5 for β=250 reconstructions, 3.2±0.6 for β=350 reconstructions, 4.1±0.6 for β=500 reconstructions, 4.7±0.5 for β=750 reconstructions and 4.8±0.5 for β=1000 reconstructions. The mean±SD scores for 68Ga-PSMA-11 PET images were 3.3±0.9 for β=250 reconstructions, 4.2±0.9 for β=350 reconstructions, 4.7±0.6 for β=500 reconstructions, 4.9±0.3 for β=750 reconstructions and 4.9±0.4 for β=1000 reconstructions. The relative observed agreement among readers for the β values of 250, 350, 500, 750, 1000 reconstructions was 49%, 50%, 60%, 70% and 74%, respectively. A total of 24 lesions (6 on RM2 and 18 on PSMA-11 scans) were detected and the mean SUVmax measurements were as follows: 13.1, 12.5, 10.4, 9.3 and 8 for the 68Ga-RM2 β values of 250, 350, 500, 750 and 1000, respectively; 22.6, 21.2, 19.7, 18.9 and 16.8 for the 68Ga-PSMA-11 β values of 250, 350, 500, 750 and 1000, respectively. Conclusions: Our results indicate that various β values should be tested and the final one selected based on consensus scoring of image quality for different 68Ga-labeled radiopharmaceuticals such as those targeting GRPR and PSMA receptors in prostate cancer. Once selected, the same β value should be consistently used since SUVmax measurements differ with different β values. %U