TY - JOUR A1 - Kazuhino, Koshino A1 - Werner, Rudolf A. A1 - Toriumi, Fuijo A1 - Javadi, Mehrbod S. A1 - Pomper, Martin G. A1 - Solnes, Lilja B. A1 - Verde, Franco A1 - Higuchi, Takahiro A1 - Rowe, Steven P. T1 - Generative Adversarial Networks for the Creation of Realistic Artificial Brain Magnetic Resonance Images JF - Tomography N2 - 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. KW - AI KW - Magnetresonanztomografie KW - artificial intelligence KW - magnetic resonance imaging KW - MRI KW - DCGAN KW - GAN KW - stroke KW - machine learning Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-172185 VL - 4 IS - 4 ER - TY - CHAP A1 - Werner, Rudolf A. A1 - Marcus, Charles A1 - Sheikhbahaei, Sara A1 - Higuchi, Takahiro A1 - Solnes, Lilja B. A1 - Rowe, Steven P. A1 - Buck, Andreas K. A1 - Lapa, Constantin A1 - Javadi, Mehrbod S. T1 - Diagnostic Accuracy of Visual Assessment of an Initial DaT-Scan in Comparison to a Fully Automatic Semiquantitative Method T2 - Journal of Nuclear Medicine N2 - No abstract available. KW - Parkinson-Krankheit KW - SPECT KW - Parkinson KW - Parkinson Disease KW - DaTscan KW - Ioflupane KW - molecular imaging Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-162208 UR - http://jnm.snmjournals.org/content/59/supplement_1/626.abstract SN - 0161-5505 N1 - This research was originally published in JNM. Rudolf A. Werner, Charles Marcus, Sara Sheikhbahaei, Takahiro Higuchi, Lilja B. Solnes, Steven P. Rowe, Andreas K. Buck, Constantin Lapa, Mehrbod S. Javadi. Diagnostic Accuracy of Visual Assessment of an Initial DaT-Scan in Comparison to a Fully Automatic Semiquantitative Method. J Nucl Med. May 1, 2018; vol. 59 no. supplement 1:626. © SNMMI. VL - 59 IS - Supplement No. 1 SP - 626 ER - TY - CHAP A1 - Werner, Rudolf A. A1 - Marcus, Charles A1 - Sheikhbahaei, Sara A1 - Higuchi, Takahiro A1 - Solnes, Lilja B. A1 - Rowe, Steven P. A1 - Buck, Andreas K. A1 - Lapa, Constantin A1 - Javadi, Mehrbod S. T1 - The Impact of Ageing on Dopamine Transporter Imaging T2 - Journal of Nuclear Medicine N2 - No abstract available. KW - Parkinson-Krankheit KW - Parkinson KW - Parkinson Disease KW - DaTscan KW - Ioflupane KW - SPECT KW - molecular imaging KW - ageing Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-162213 UR - http://jnm.snmjournals.org/content/59/supplement_1/1646.abstract SN - 0161-5505 N1 - This research was originally published in JNM. Rudolf A. Werner, Charles Marcus, Sara Sheikhbahaei, Takahiro Higuchi, Lilja B. Solnes, Steven P. Rowe, Andreas K. Buck, Constantin Lapa, Mehrbod S. Javadi. The Impact of Ageing on Dopamine Transporter Imaging. J Nucl Med. May 1, 2018; vol. 59 no. supplement 1:1646. © SNMMI. VL - 59 IS - Supplement No 1 SP - 1646 ER - TY - JOUR A1 - Werner, Rudolf A. A1 - Marcus, Charles A1 - Sheikhbahaei, Sara A1 - Solnes, Lilja B. A1 - Leal, Jeffrey P. A1 - Du, Yong A1 - Rowe, Steven P. A1 - Higuchi, Takahiro A1 - Buck, Andreas K. A1 - Lapa, Constantin A1 - Javadi, Mehrbod S. T1 - Visual and Semiquantitative Accuracy in Clinical Baseline 123I-Ioflupane SPECT/CT Imaging JF - Clinical Nuclear Medicine N2 - PURPOSE: We aimed to (a) elucidate the concordance of visual assessment of an initial I-ioflupane scan by a human interpreter with comparison to results using a fully automatic semiquantitative method and (b) to assess the accuracy compared to follow-up (f/u) diagnosis established by movement disorder specialists. METHODS: An initial I-ioflupane scan was performed in 382 patients with clinically uncertain Parkinsonian syndrome. An experienced reader performed a visual evaluation of all scans independently. The findings of the visual read were compared with semiquantitative evaluation. In addition, available f/u clinical diagnosis (serving as a reference standard) was compared with results of the human read and the software. RESULTS: When comparing the semiquantitative method with the visual assessment, discordance could be found in 25 (6.5%) of 382 of the cases for the experienced reader (ĸ = 0.868). The human observer indicated region of interest misalignment as the main reason for discordance. With neurology f/u serving as reference, the results of the reader revealed a slightly higher accuracy rate (87.7%, ĸ = 0.75) compared to semiquantification (86.2%, ĸ = 0.719, P < 0.001, respectively). No significant difference in the diagnostic performance of the visual read versus software-based assessment was found. CONCLUSIONS: In comparison with a fully automatic semiquantitative method in I-ioflupane interpretation, human assessment obtained an almost perfect agreement rate. However, compared to clinical established diagnosis serving as a reference, visual read seemed to be slightly more accurate as a solely software-based quantitative assessment. KW - Single-Photon-Emissions-Computertomographie KW - SPECT KW - Parkinson’s disease KW - Parkinsonism KW - DaTscan KW - 123I-Ioflupane KW - SPECT KW - SPECT/CT Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-168181 SN - 1536-0229 VL - 44 IS - 1 ER - TY - JOUR A1 - Werner, Rudolf A. A1 - Sheikhbahaei, Sara A1 - Jones, Krystyna M. A1 - Javadi, Mehrbod S. A1 - Solnes, Lilja B. A1 - Ross, Ashley E. A1 - Allaf, Mohamad E. A1 - Pienta, Kenneth J. A1 - Lapa, Constantin A1 - Buck, Andreas K. A1 - Higuchi, Takahiro A1 - Pomper, Martin G. A1 - Gorin, Micheal A. A1 - Rowe, Steven P. T1 - Patterns of uptake of prostate-specific membrane antigen (PSMA)-targeted \(^{18}\)F-DCFPyL in peripheral ganglia JF - Annals of Nuclear Medicine N2 - 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. KW - 18F-DCFPL KW - Positronen-Emissions-Tomografie KW - Prostata KW - PSMA KW - Ganglia KW - Pitfall KW - PET KW - Tracer KW - Radiotracer KW - Imaging pitfalls KW - Prostate Cancer Y1 - 2017 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-166971 SN - 0914-7187 VL - 31 IS - 9 ER - TY - JOUR A1 - Werner, Rudolf A. A1 - Weich, Alexander A1 - Kircher, Malte A1 - Solnes, Lilja B. A1 - Javadi, Mehrbod S. A1 - Higuchi, Takahiro A1 - Buck, Andreas K. A1 - Pomper, Martin G. A1 - Rowe, Steven A1 - Lapa, Constantin T1 - The theranostic promise for neuroendocrine tumors in the late 2010s – Where do we stand, where do we go? JF - Theranostics N2 - More than 25 years after the first peptide receptor radionuclide therapy (PRRT), the concept of somatostatin receptor (SSTR)-directed imaging and therapy for neuroendocrine tumors (NET) is seeing rapidly increasing use. To maximize the full potential of its theranostic promise, efforts in recent years have expanded recommendations in current guidelines and included the evaluation of novel theranostic radiotracers for imaging and treatment of NET. Moreover, the introduction of standardized reporting framework systems may harmonize PET reading, address pitfalls in interpreting SSTR-PET/CT scans and guide the treating physician in selecting PRRT candidates. Notably, the concept of PRRT has also been applied beyond oncology, e.g. for treatment of inflammatory conditions like sarcoidosis. Future perspectives may include the efficacy evaluation of PRRT compared to other common treatment options for NET, novel strategies for closer monitoring of potential side effects, the introduction of novel radiotracers with beneficial pharmacodynamic and kinetic properties or the use of supervised machine learning approaches for outcome prediction. This article reviews how the SSTR-directed theranostic concept is currently applied and also reflects on recent developments that hold promise for the future of theranostics in this context. KW - theranostics KW - Positronen-Emissions-Tomografie KW - PRRT KW - somatostatin receptor KW - peptide receptor radionuclide therapy KW - neuroendocrine tumor Y1 - 2018 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-170264 VL - 8 IS - 22 ER -