Organ Localization in PET/CT Images using Hierarchical Conditional Faster R-CNN Method

Abstract Background [18F]Fluoromisonidazole ([18F]FMISO) is a PET imaging probe widely used for the detection of hypoxia. We previously reported that [18F]FMISO is metabolized to the glutathione conjugate of the reduced form in hypoxic cells. In addition, we found that the [18F]FMISO uptake level varied depending on the cellular glutathione conjugation and excretion ability such as enzyme activity of glutathione-S-transferase and expression levels of multidrug resistance-associated protein 1 (MRP1, an efflux transporter), in addition to the cellular hypoxic state. In this study, we evaluated whether MRP1 activity affected [18F]FMISO PET imaging. Methods FaDu human pharyngeal squamous cell carcinoma cells were pretreated with MRP1 inhibitors (cyclosporine A, lapatinib, or MK-571) for 1 h, incubated with [18F]FMISO for 4 h under hypoxia, and their radioactivity was then measured. FaDu tumor-bearing mice were intravenously injected with [18F]FMISO, and PET/CT images were acquired at 4 h post-injection (1st PET scan). Two days later, the same mice were pretreated with MRP1 inhibitors (cyclosporine A, lapatinib, or MK-571) for 1 h, and PET/CT images were acquired (2nd PET scan). Results FaDu cells pretreated with MRP1 inhibitors exhibited significantly higher radioactivity than those without inhibitor treatment (cyclosporine A: 6.91 ± 0.27, lapatinib: 10.03 ± 0.47, MK-571: 10.15 ± 0.44%dose/mg protein, p < 0.01). In the in vivo PET study, the SUVmean ratio in tumors [calculated as after treatment (2nd PET scan)/before treatment of MRP1 inhibitors (1st PET scan)] of the mice treated with MRP1 inhibitors was significantly higher than those of control mice (cyclosporine A: 2.6 ± 0.7, lapatinib: 2.2 ± 0.7, MK-571: 2.2 ± 0.7, control: 1.2 ± 0.2, p < 0.05). Conclusion In this study, we revealed that MRP1 inhibitors increase [18F]FMISO accumulation in hypoxic cells. This suggests that [18F]FMISO-PET imaging is affected by MRP1 inhibitors independent of the hypoxic state.

Download Full-text

Impact of the Q.Clear reconstruction algorithm on the interpretation of PET/CT images in patients with lymphoma

EJNMMI Research ◽

10.1186/s13550-020-00690-6 ◽

2020 ◽

Vol 10 (1) ◽

Author(s):

Michał Wyrzykowski ◽

Natalia Siminiak ◽

Maciej Kaźmierczak ◽

Marek Ruchała ◽

Rafał Czepczyński

Keyword(s):

Reconstruction Algorithm ◽

Ct Images ◽

Pet Ct

Download Full-text

Automated quantification of baseline imaging PET metrics on FDG PET/CT images of pediatric Hodgkin lymphoma patients

EJNMMI Physics ◽

10.1186/s40658-020-00346-3 ◽

2020 ◽

Vol 7 (1) ◽

Author(s):

Amy J. Weisman ◽

Jihyun Kim ◽

Inki Lee ◽

Kathleen M. McCarten ◽

Sandy Kessel ◽

...

Keyword(s):

Hodgkin Lymphoma ◽

Correlation Coefficient ◽

Pet Imaging ◽

Fdg Pet ◽

Relative Difference ◽

Ct Images ◽

Imaging Features ◽

Pediatric Hodgkin Lymphoma ◽

Pet Ct ◽

Fdg Pet Ct

Abstract Purpose For pediatric lymphoma, quantitative FDG PET/CT imaging features such as metabolic tumor volume (MTV) are important for prognosis and risk stratification strategies. However, feature extraction is difficult and time-consuming in cases of high disease burden. The purpose of this study was to fully automate the measurement of PET imaging features in PET/CT images of pediatric lymphoma. Methods 18F-FDG PET/CT baseline images of 100 pediatric Hodgkin lymphoma patients were retrospectively analyzed. Two nuclear medicine physicians identified and segmented FDG avid disease using PET thresholding methods. Both PET and CT images were used as inputs to a three-dimensional patch-based, multi-resolution pathway convolutional neural network architecture, DeepMedic. The model was trained to replicate physician segmentations using an ensemble of three networks trained with 5-fold cross-validation. The maximum SUV (SUVmax), MTV, total lesion glycolysis (TLG), surface-area-to-volume ratio (SA/MTV), and a measure of disease spread (Dmaxpatient) were extracted from the model output. Pearson’s correlation coefficient and relative percent differences were calculated between automated and physician-extracted features. Results Median Dice similarity coefficient of patient contours between automated and physician contours was 0.86 (IQR 0.78–0.91). Automated SUVmax values matched exactly the physician determined values in 81/100 cases, with Pearson’s correlation coefficient (R) of 0.95. Automated MTV was strongly correlated with physician MTV (R = 0.88), though it was slightly underestimated with a median (IQR) relative difference of − 4.3% (− 10.0–5.7%). Agreement of TLG was excellent (R = 0.94), with median (IQR) relative difference of − 0.4% (− 5.2–7.0%). Median relative percent differences were 6.8% (R = 0.91; IQR 1.6–4.3%) for SA/MTV, and 4.5% (R = 0.51; IQR − 7.5–40.9%) for Dmaxpatient, which was the most difficult feature to quantify automatically. Conclusions An automated method using an ensemble of multi-resolution pathway 3D CNNs was able to quantify PET imaging features of lymphoma on baseline FDG PET/CT images with excellent agreement to reference physician PET segmentation. Automated methods with faster throughput for PET quantitation, such as MTV and TLG, show promise in more accessible clinical and research applications.

Download Full-text