Refine
Has Fulltext
- yes (6)
Is part of the Bibliography
- yes (6)
Document Type
- Journal article (4)
- Conference Proceeding (2)
Language
- English (6)
Keywords
- DaTscan (3)
- SPECT (3)
- Ioflupane (2)
- Parkinson (2)
- Parkinson Disease (2)
- Parkinson-Krankheit (2)
- Positronen-Emissions-Tomografie (2)
- molecular imaging (2)
- 123I-Ioflupane (1)
- 18F-DCFPL (1)
Institute
- Klinik und Poliklinik für Nuklearmedizin (6) (remove)
Sonstige beteiligte Institutionen
EU-Project number / Contract (GA) number
- 701983 (6)
Objective: Radiotracers targeting prostate-specific membrane antigen (PSMA) have increasingly been recognized as showing uptake in a number of normal structures, anatomic variants, and non-prostate-cancer pathologies. We aimed to explore the frequency and degree of uptake in peripheral ganglia in patients undergoing PET with the PSMA-targeted agent \(^{18}\)F-DCFPyL.
Methods: A total of 98 patients who underwent \(^{18}\)F-DCFPyL PET/CT imaging were retrospectively analyzed. This included 76 men with prostate cancer (PCa) and 22 patients with renal cell carcinoma (RCC; 13 men, 9 women). Scans were evaluated for uptake in the cervical, stellate, celiac, lumbar and sacral ganglia. Maximum standardized uptake value corrected to body weight (SUV\(_{max}\)), and maximum standardized uptake value corrected to lean body mass (SUL\(_{max}\)) were recorded for all ganglia with visible uptake above background. Ganglia-to-background ratios were calculated by dividing the SUV\(_{max}\) and SUL\(_{max}\) values by the mean uptake in the ascending aorta (Aortamean) and the right gluteus muscle (Gluteusmean).
Results: Overall, 95 of 98 (96.9%) patients demonstrated uptake in at least one of the evaluated peripheral ganglia. With regard to the PCa cohort, the most frequent sites of radiotracer accumulation were lumbar ganglia (55/76, 72.4%), followed by the cervical ganglia (51/76, 67.1%). Bilateral uptake was found in the majority of cases [lumbar 44/55 (80%) and cervical 30/51 (58.8%)]. Additionally, discernible radiotracer uptake was recorded in 50/76 (65.8%) of the analyzed stellate ganglia and in 45/76 (59.2%) of the celiac ganglia, whereas only 5/76 (6.6%) of the sacral ganglia demonstrated \(^{18}\)F-DCFPyL accumulation. Similar findings were observed for patients with RCC, with the most frequent locations of radiotracer uptake in both the lumbar (20/22, 90.9%) and cervical ganglia (19/ 22, 86.4%). No laterality preference was found in mean PSMA-ligand uptake for either the PCa or RCC cohorts.
Conclusion: As PSMA-targeted agents become more widely disseminated, the patterns of uptake in structures that are not directly relevant to patients’ cancers must be understood. This is the first systematic evaluation of the uptake of \(^{18}\)F-DCFPyL in ganglia demonstrating a general trend with a descending frequency of radiotracer accumulation in lumbar, cervical, stellate, celiac, and sacral ganglia. The underlying biology that leads to variability of PSMA-targeted radiotracers in peripheral ganglia is not currently understood, but may provide opportunities for future research.
Even as medical data sets become more publicly accessible, most are restricted to specific medical conditions. Thus, data collection for machine learning approaches remains challenging, and synthetic data augmentation, such as generative adversarial networks (GAN), may overcome this hurdle. In the present quality control study, deep convolutional GAN (DCGAN)-based human brain magnetic resonance (MR) images were validated by blinded radiologists. In total, 96 T1-weighted brain images from 30 healthy individuals and 33 patients with cerebrovascular accident were included. A training data set was generated from the T1-weighted images and DCGAN was applied to generate additional artificial brain images. The likelihood that images were DCGAN-created versus acquired was evaluated by 5 radiologists (2 neuroradiologists [NRs], vs 3 non-neuroradiologists [NNRs]) in a binary fashion to identify real vs created images. Images were selected randomly from the data set (variation of created images, 40%-60%). None of the investigated images was rated as unknown. Of the created images, the NRs rated 45% and 71% as real magnetic resonance imaging images (NNRs, 24%, 40%, and 44%). In contradistinction, 44% and 70% of the real images were rated as generated images by NRs (NNRs, 10%, 17%, and 27%). The accuracy for the NRs was 0.55 and 0.30 (NNRs, 0.83, 0.72, and 0.64). DCGAN-created brain MR images are similar enough to acquired MR images so as to be indistinguishable in some cases. Such an artificial intelligence algorithm may contribute to synthetic data augmentation for "data-hungry" technologies, such as supervised machine learning approaches, in various clinical applications.