Automated F18-FDG PET/CT image quality assessment using deep neural networks on a latest 6-ring digital detector system.

To evaluate whether a machine learning classifier can evaluate image quality of maximum intensity projection (MIP) images from F18-FDG-PET scans. A total of 400 MIP images from F18-FDG-PET with simulated decreasing acquisition time (120 s, 90 s, 60 s, 30 s and 15 s per bed-position) using block sequential regularized expectation maximization (BSREM) with a beta-value of 450 and 600 were created. A machine learning classifier was fed with 283 images rated "sufficient image quality" and 117 images rated "insufficient image quality". The classification performance of the machine learning classifier was assessed by calculating sensitivity, specificity, and area under the receiver operating characteristics curve (AUC) using reader-based classification as the target. Classification performance of the machine learning classifier was AUC 0.978 for BSREM beta 450 and 0.967 for BSREM beta 600. The algorithm showed a sensitivity of 89% and 94% and a specificity of 94% and 94% for the reconstruction BSREM 450 and 600, respectively. Automated assessment of image quality from F18-FDG-PET images using a machine learning classifier provides equivalent performance to manual assessment by experienced radiologists.

Histopathological Features of Parathyroid Adenoma and 18F-Choline Uptake in PET/MR of Primary Hyperparathyroidism.

BACKGROUND: The aim of this study was to assess the relationship between the histopathological properties of hyperfunctioning parathyroids and parathyroid 18F-choline uptake. PATIENTS AND METHODS: A total of 31 parathyroid adenomas were retrospectively analyzed in patients with primary hyperparathyroidism and preoperative 18F-choline PET/MR. PET/MR parameters of parathyroid glands (SUVmax and target-to-background ratio in early-phase [EP] and late-phase [LP]), MRI volume, preoperative parathyroid hormone (PTH) serum concentration, and postoperative histopathology (predominant cell type and growth pattern of adenoma cells, location and size of adenoma) were assessed. The relationship of PET/MR parameters, PTH, and histological parameters was determined using linear regression, Spearman correlation and Kruskal-Wallis test. RESULTS: The median volume of parathyroid adenoma was 421.78 ± 142.46 mm3 (46.39-4412.69). Adenomas were predominantly composed of chief, water-clear, and oncocytic/oxyphilic cells in 27/31, 2/31, and 2/31 cases, respectively. The growth pattern was predominantly solid, follicular, and trabecular in 18/31, 8/31, and 5/31, respectively. The SUVmax was 6.71 ± 3.39 in EP and 6.91 ± 3.97 in LP. Follicular growth pattern had slightly higher EP SUVmax (trabecular: 4.12 ± 0.56; solid: 6.62 ± 3.19; follicular: 8.56 ± 3.96; P = 0.046). Spearman correlation showed strong positive correlation between volume and both EP and LP SUVmax (0.626; P = 0.0001 and 0.576; P = 0.0001, respectively). Linear regression analysis revealed significant correlation between PTH level and EP and LP SUVmax (both P = 0.001); in contrast, no correlation was found between PTH level and both cell type and growth pattern. CONCLUSIONS: Our findings suggest that 18F-choline uptake of parathyroid adenomas might be associated both with the histological growth pattern and adenoma volume, but not with a specific cell type.

[18F]-sodium fluoride PET/MR for painful lumbar facet joint degeneration - a randomized controlled clinical trial.

BACKGROUND CONTEXT: [18F]-sodium fluoride (NaF) PET/MR is a modern diagnostic modality for imaging increased bone turnover. Its merits in detecting painful facet joint osteoarthritis in patients with lumbar back pain are unknown. PURPOSE: To perform a prospective randomized controlled study investigating [18F]-NaF PET/MR for detecting painful facet joints in comparison to the standard of care (SOC), including clinical examination and conventional MRI. STUDY DESIGN/SETTING: Randomized controlled clinical study. PATIENT SAMPLE: Thirty-nine patients. OUTCOME MEASURES: Visual analog pain scale (VAS) before and at several time points after facet joint infiltration. METHODS: Patients with low back pain and suspected facet joint osteoarthritis underwent lumbar [18F]-NaF PET/MR, besides conventional MRI and clinical examination. After randomization, they either received local anesthetics/ corticosteroid infiltration of facet joints as defined by clinical examination and conventional MRI (SOC), or according to the hot spots on PET/MR. VAS was documented at 15 minutes, 1 day, 1 week and 1 month after infiltration. Thirty-nine patients underwent PET/MR before the study was stopped due to new Good Manufacturing Practice requirement and new regulations by radiation protection authorities limiting staff radiation exposure during the production of this radiotracer. RESULTS: Significant pain reduction compared to baseline was shown at every timepoint in both groups, except at 1 month after infiltration in the SOC group. Pain levels did not differ between SOC (n=17) and PET/MR patients (n=12) before infiltration and at 15 minutes, 1 day, 1 week and 1 month after infiltration. No significant correlation was detected between the sum of the PET/MR activity and the initial pain scores or relative reduction of pain after 15 minutes. The constructed study groups of patients with infiltration of all facet joints being PET/MR-positive (n=18) had significantly less pain after 1 months than patients with infiltration in PET/MR-negative facet joints (n=11) (VAS: 4 [0, 9] vs. 7 [2, 10], p=.046). CONCLUSIONS: There is no correlation of pain to NaF activity nor a relevant superiority of [18F]-NaF PET/MR for identification of painful facet joints compared to the standard of care.

Attenuation Correction Using Template PET Registration for Brain PET: A Proof-of-Concept Study.

NeuroLF is a dedicated brain PET system with an octagonal prism shape housed in a scanner head that can be positioned around a patient's head. Because it does not have MR or CT capabilities, attenuation correction based on an estimation of the attenuation map is a crucial feature. In this article, we demonstrate this method on [18F]FDG PET brain scans performed with a low-resolution proof of concept prototype of NeuroLF called BPET. We perform an affine registration of a template PET scan to the uncorrected emission image, and then apply the resulting transform to the corresponding template attenuation map. Using a whole-body PET/CT system as reference, we quantitively show that this method yields comparable image quality (0.893 average correlation to reference scan) to using the reference µ-map as obtained from the CT scan of the imaged patient (0.908 average correlation). We conclude from this initial study that attenuation correction using template registration instead of a patient CT delivers similar results and is an option for patients undergoing brain PET.