Search

Abstract Background Prostate-specific membrane antigen (PSMA) PET/CT imaging is widely used for quantitative image analysis, especially in radioligand therapy (RLT) for metastatic castration-resistant prostate cancer (mCRPC). Unknown features influencing PSMA biodistribution can be explored by analyzing segmented organs at risk (OAR) and lesions. Manual segmentation is time-consuming and labor-intensive, so automated segmentation methods are desirable. Training deep-learning segmentation models is challenging due to the scarcity of high-quality annotated images. Addressing this, we developed shifted windows UNEt TRansformers (Swin UNETR) for fully automated segmentation. Within a self-supervised framework, the model’s encoder was pre-trained on unlabeled data. The entire model was fine-tuned, including its decoder, using labeled data. Methods In this work, 752 whole-body [68Ga]Ga-PSMA-11 PET/CT images were collected from two centers. For self-supervised model pre-training, 652 unlabeled images were employed. The remaining 100 images were manually labeled for supervised training. In the supervised training phase, 5-fold cross-validation was used with 64 images for model training and 16 for validation, from one center. For testing, 20 hold-out images, evenly distributed between two centers, were used. Image segmentation and quantification metrics were evaluated on the test set compared to the ground-truth segmentation conducted by a nuclear medicine physician. Results The model generates high-quality OARs and lesion segmentation in lesion-positive cases, including mCRPC. The results show that self-supervised pre-training significantly improved the average dice similarity coefficient (DSC) for all classes by about 3%. Compared to nnU-Net, a well-established model in medical image segmentation, our approach outperformed with a 5% higher DSC. This improvement was attributed to our model’s combined use of self-supervised pre-training and supervised fine-tuning, specifically when applied to PET/CT input. Our best model had the lowest DSC for lesions at 0.68 and the highest for liver at 0.95. Conclusions We developed a state-of-the-art neural network using self-supervised pre-training on whole-body [68Ga]Ga-PSMA-11 PET/CT images, followed by fine-tuning on a limited set of annotated images. The model generates high-quality OARs and lesion segmentation for PSMA image analysis. The generalizable model holds potential for various clinical applications, including enhanced RLT and patient-specific internal dosimetry.

This paper contains single-center prospective information showing illustrative examples of chemokine receptor-4 (CXCR4) targeting in high-grade glial brain tumors in treatment-naïve adult patients using a novel radiolabeled PET tracer: [68Ga]Ga-CXCR4 PET/CT.High-grade glioma is one of the most resistant malignancies to treatment. Despite major breakthroughs in diagnostic and therapeutic approaches, the overall 5-year survival rate remains in the 5–10% range. CXCR4 is a chemokine with the C-X-C motif that is overexpressed in high-grade gliomas.The 24 consecutive treatment- naïve enrolled patients underwent PET/CT images using the SIEMENS scanner (Biograph6 TrueV) and received the radiotracer intravenously. After approximately 60 min, the PET/CT acquisition was performed with a dedicated scanner and in 10 min time per bed position. The images were reconstructed and analyzed with the 3D-OSEM algorithm, applying point spread function (PSF) or resolution recovery algorithm (TrueX in Syngo ® software, Siemens Medical Solution), 3 iterations, and 21 subsets using a 3 mm Gaussian post-smoothing filter.These data would be potentially beneficial for automatic tumor delineation machine learning after augmented with other data retrieved from different papers as well as for differentiation between an active viable tumor vs. post-surgery/necrosis in indeterminate cases. The theranostics potential (CXCR4-tageted labeled beta emitters) is one of the most novel areas of interest for future studies.

Objective(s): Nowadays, ionizing radiation (IR) has a significant contribution to the diagnostic and therapeutic medicine, and following that, health risks to individuals through unexpected exposure is greatly increased. Therefore, biological and molecular technology for estimation of dose (biodosimetry) is taken into consideration. In biodosimetry methods stimulation of cells to proliferation is routine to achieve more sensitivity of techniques. However, this concept has recently been challenged by new molecular methods such as gene expression analysis. This study aims to investigate the stimulation effects on gene expression biodosimetry.Materials and Methods: The blood samples were taken from15 patients who were irradiated by TC-99 MIBI, before radiopharmaceutical injection and 24 hr after injection. Lymphocytes were extracted immediately and activated by (phytohemagglutinin) PHA for 24 hr and XPA and FDXR expression levels were investigated by employing relative quantitative Real-Time PCR.Results: The results of this study show a significant increase in the FDXR expression level and a significant decrease in the XPA after stimulation of irradiated lymphocytes. Interestingly, a significant increasing trend in the FDXR expression level (at 0.05 significance level) following cell stimulation to the division was observed.Conclusion: Our results suggest that the PHA activation role in gene expression-based biodosimetry is strongly depended on the target genes and the relevant protein pathways. Finally, cell stimulation looks to be useful for some specific genes, such as FDXR, due to the increasing trend in expression and improvement of sensitivity of gene expression-based biodosimetry method.

The data presented here provide information about the role of reconstruction parameters on Positron Emission Tomography (PET) image quantification. Multiple phantom measurements in four different Spheres to Background Ratio (SBR) were performed on Biograph 6 TruePoint TrueV PET/CT scanner. PET raw data were reconstructed with/without resolution recovery algorithm using six various iteration x subsets with five different Full-Width Half-Maximum (FWHM) values of Gaussian post-smoothing filter. The Recovery Coefficient (RC) of six spheres using three common Volume of Interest (VOI) methods (max, 3D-50% Isocontour, and peak) were calculated. Moreover, SUVmax, SUVmean, and SUVpeak and volumetric indices, such as metabolic tumor volume (MTV), volume recovery coefficient (VRC), and total lesion glycolysis (TLG) were measured. RCmax, RC50%, and RCpeak as a function of sphere size were plotted in all reconstruction methods considering different SBRs. The data could be noticeable for standardization and optimization of quantitative metrics in PET imaging.

Objective(s): In recent years, the application of radiopharmaceuticals in nuclear medicine has increased substantially. Following the diagnostic procedures performed in nuclear medicine departments, such as myocardial perfusion imaging, patients generally receive considerable doses of radiation. Normally, radiation-induced DNA damages are expected following exposure to a low-dose ionizing radiation. In order to detect molecular changes, high-sensitivity techniques must be utilized. The aim of this study was to assess the effect of a low-dose (below 10 mSv) gamma ray on gene expression using quantitative real-time polymerase chain reaction (qRT-PCR). Methods: Blood samples were obtained from 20 volunteer patients who underwent myocardial perfusion imaging. They were given various doses of Technetium99-m methoxyisobutylisonitrile (99mTc-MIBI). After that, peripheral blood mononuclear cells (PBMNs) were derived, and then total RNA was extracted and reverse-transcribed to cDNA. Finally, the expression levels of xeroderma pigmentosum complementation group-A (XPA) and ferredoxin reductase (FDXR) genes were determinded through qRT-PCR technique using SYBR Green. Results: XPA and FDXR expression levels were obtained following a very low-dose ionizing radiation. A significant up-regulation of both genes was observed, and the gene expression level of each individual patient was different. If differences in the administered activity and radiosensitivity are taken into account, the observed differences could be justified. Furthermore, gender and age did not play a significant role in the expression levels of the genes under study. Conclusion: The up-regulation of FDXR after irradiation revealed the high-sensitivity level of this gene; therefore, it could be used as an appropriate biomarker for biological dosimetry. On the other hand, the up-regulation of XPA is an indication of DNA repair following radiation exposure. According to linear no-threshold model (LNT) and the results obtained from this study, a very low dose of ionizing radiation could bring about adverse biological effects at molecular level in the irradiated person.