Purpose: This study aimed to evaluate the image quality-dose trade-off in pregnant patients imaged with long-axial field-of-view [18F]FDG PET/CT and to identify the most predictive body composition metric for image quality to develop a pregnancy-tailored dosage model.
Methods: Patients imaged with [18F]FDG PET/CT according to local pregnancy protocols were included in this study. Using raw PET data, images of various degrees of image quality were reconstructed. Acceptable image quality was identified using signal-to-noise ratio (SNR) in the liver and Likert scores. The minimum required scan statistics was modelled based on SNR and patient body composition. F-tests were used to find the best-fitting model parameter out of weight, weight-to-height-ratio, body-mass-index, and body surface area (BSA). Foetal dose was estimated with PET conversion factors and size-specific CT dose index values.
Results: Eleven patients were included in image quality analysis and dosage model optimization. SNR strongly correlated with Likert scores (R² = 0.80), with 10.72 SNR indicating acceptable image quality. BSA best predicted image quality (R² = 0.85), outperforming weight (R² = 0.78), weight-to-height ratio (R² = 0.63), and body mass index (R² = 0.38). The proposed dosage model reduces activity by 41-96% compared to current local pregnancy and adult protocols, resulting in estimated foetal radiation doses of 0.066 mGy (PET) and 0.31 mGy (CT).
Conclusion: BSA accurately predicted [18F]FDG PET/CT image quality in pregnant patients. The proposed dosage regimen allows significant dose reduction and can be used as a foundation for the development of pregnancy dosage protocols.
Background: Individualized radiopharmaceutical therapies guided by patient-specific absorbed dose assessments using imaging have the potential to improve both efficacy and safety. Understanding sources of variability in absorbed dose calculations is critical for standardization. The Society of Nuclear Medicine and Molecular Imaging Dosimetry Task Force launched the 177Lu Dosimetry Challenge to evaluate variability across different steps within the dosimetry workflow. This work aimed to assess the variability in absorbed doses due to differences in segmentation methods.
Methods: Anonymized datasets from two patients treated with 177Lu-DOTATATE, including serial SPECT/CT scans, were made available online. Participants were asked to segment healthy organs and lesions and perform dosimetry calculations. In a subsequent task, participants were provided with standardized segmented VOIs and asked to perform dosimetry based on these pre-defined regions. Variability in segmentation was assessed by comparing absorbed dose estimates across two scenarios: participant-generated segmentations versus predefined reference segmentations. Relative absorbed dose variability was quantified using the quartile coefficient of dispersion (QCD) and interquartile range.
Results: Variability in absorbed dose (measured as QCD difference between absorbed doses from participant-generated segmentations and those from reference segmentations) for kidneys was less than 5% in simple cases and 10.6% for more challenging scenarios (i.e. presence of intraparenchymal cysts, cortical defects). Lesion segmentation exhibited higher variability, with absorbed dose variability reaching up to 22.4%.
Conclusions: Segmentation significantly contributes to variability in absorbed dose estimates, particularly for lesions and for kidneys with anatomical complexities. Standardizing segmentation protocols and providing training on advanced segmentation methods are essential to reduce variability.
Introduction: Positron Emission Tomography (PET) imaging is a close ally of Precision Medicine, and it has been proven to be indispensable in the field of Psychiatry. This imaging modality may also present an important role in understanding Neurodevelopmental disorders and their link to Psychiatric conditions, with new highly selective binders being used currently in research. PET imaging requires the administration of radiopharmaceuticals, where the radioisotope is in incorporated into a highly selective binder. Dosimetry and injected activity optimisation play a crucial role in the field of PET imaging as they allow to determine the radiation dose absorbed by target and non-target tissues, and determine the lowest amount required to deliver images with diagnostic quality and obtain reliable quantitative data, without overexposing patients. The aim of this research is to investigate the feasibility of reducing the injected activity of the [11C]-(+)-PHNO and [11C]UCB-J radiopharmaceuticals, for patients with neurodevelopmental disorders who undergo brain imaging in the PET-Magnetic Resonance (MR) scanner, without compromising quantitative accuracy of outcome measures.
Results: No statistically significant differences were found when comparing the 1/2 to 1/6 datasets with the full injected activity [11C]-(+)-PHNO dataset. Furthermore, the findings obtained from investigating the impact of low injected activity administrations of [11C]UCB-J revealed that it is possible to reduce the administered activity by 1/2, when the clinical outcome measure under evaluation is the binding potential relative to non-displaceable volume (BPND). When the outcome measure under investigation is the standard uptake volume ratio (SUVR), it is possible to decrease the injected activity to 1/3, for [11C]UCB-J.
Conclusions: The simulation and analysis methodologies deployed in this project are suitable for investigating scans with low injected activity for tracers with cortical and striatal uptake, when the outcome measure assessed is the BPND or the SUVR. Whilst the data suggests that imaging with low injected activity is achievable, the efficacy of the investigation is highly dependent on the algorithm used to reconstruct the images, the outcome measure and the radiopharmaceutical used to acquire the PET-MR scans. For the [11C]UCB-J radiopharmaceutical, it is possible to decrease the injectable activity to 1/3 of the original administration without compromising the SUVR.
Recent advancements in nuclear medicine, particularly in personalised radiopharmaceutical therapy, have emphasised the growing need for precise assessments of therapeutic safety and efficacy. These evaluations depend heavily on individual patient pharmacokinetics and dosimetry studies. Pharmacokinetics are typically assessed using whole-body SPECT/CT or PET/CT time-point imaging, preceded by rigorous calibration procedures to ensure the accuracy of absorbed dose calculations. The growing need for reliable imaging data has driven the development and adoption of various software tools aimed at optimising the processing, analysis, and dosimetry of nuclear medicine images. Open-source solutions are increasingly bridging the gap in accessibility, especially in resource-constrained environments, while AI-driven segmentation and time-activity curve modelling are emerging as critical innovations for improving workflow efficiency. Future efforts should prioritise validation, standardisation, and the development of robust tools tailored to complex dosimetry scenarios, including alpha and Auger therapies. This review evaluates several available software tools, both open-source and commercial, for processing calibration phantoms and patient images with an emphasis on quantitative analyses. It also examines tools used for post-imaging dosimetry. Key advancements in computational techniques are highlighted, including algorithms for dose calculation, computational models, and applications in deep learning. Furthermore, the review addresses existing limitations and ongoing efforts to enhance the accuracy, reproducibility, and clinical integration of these technologies. Future directions include integration of ultra-high-sensitivity detectors (e.g., long-axial-FOV PET and full-ring SPECT), wider adoption of standardised reconstruction and quantification workflows, incorporation of targeted alpha therapy and Auger models, improved uncertainty propagation, and routine implementation of accelerated clinical dosimetry pipelines. This manuscript aims to help support researchers, medical physicists, and clinicians in effectively adopting and applying these tools to improve outcomes in nuclear medicine practices.
Ischemic heart disease remains a leading cause of mortality worldwide. Myocardial perfusion imaging (MPI) using Rubidium-82 (82Rb) positron emission tomography (PET) is a cornerstone in its evaluation. However, conventional CT-based attenuation correction (AC) is prone to artifacts, with misalignment between PET emission data and the CT-AC being a common problem. This study evaluates the feasibility of introducing a deep learning approach to generate synthetic CT (sCT) images directly from non-attenuation-corrected 82Rb-PET images. To this end, we developed a cGAN using a conditional generative adversarial network (cGAN) with an Attention U-Net generator to produce sCT images to produce sCT-AC maps, based upon 544 PET/CT MPI scans. Image quality was assessed using structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), mean absolute error (MAE), and mean error (ME). Additionally, attenuation-corrected PET images based on sCT were evaluated in the cardiac region using relative mean error (RME) and relative mean absolute error (RMAE). Cardiac function and perfusion assessments, defined as the ischemic total perfusion deficit (iTPD) and the left ventricular ejection fraction reserve (LVEFR), were compared between sCT-based and conventional CT AC methods. Our sCT-images provided good correlation to the conventional CT-AC (SSIM = 0.91 ± 0.037, PSNR = 29.9 ± 3.2 dB). For the PET images, we report a slight bias in the cardiac region (RME = 4.2 ± 7.8%, RMAE = 6.9 ± 5.9%), likely due to a uniform overestimation of the soft-tissue u-maps within the sCT. Despite the bias, the quantification metrics remained comparable to those obtained with CT AC (mean iTPD: CT 3.73 ± 5.19% vs. sCT 3.67 ± 5.13%; mean LVEFR: CT 5.88 ± 5.96% vs. sCT 5.90 ± 6.11%). Additionally, the sCT-based approach appeared to reduce motion and implant-related artifacts, providing further motivation for its use over CT. This observation was made through visual inspection on a case-by-case basis. These results demonstrate the potential of deep learning-based sCT generation to maintain integrity in PET MPI while helping to mitigate issues related to misalignments and metal-induced artifacts.
Background: Internal mammary lymph node (IMLN) metastases play an important role in breast cancer staging and treatment planning but is often difficult to detect because of their small size and anatomical location. Recent advances in digital time-of-flight (TOF) positron emission tomography (PET)/CT and advanced image reconstruction techniques may improve the visualization of such small lesions. This study aimed to evaluate the performance of advanced reconstruction methods (HYPER Iterative and uAI HYPER DPR) for visualizing IMLN metastases in breast cancer using phantom and clinical data.
Methods: A modified NEMA image quality phantom and a retrospective cohort of breast cancer patients with IMLN metastases were evaluated using a high-resolution digital TOF PET/CT system (uMI 550). Images were reconstructed using ordered subset expectation maximization (OSEM), HYPER Iterative, and uAI HYPER DPR with different reconstruction parameters, and quantitative metrics and visual scores were assessed.
Results: In both phantom and clinical images, smaller RS-values for HYPER Iterative and larger Str-values for uAI HYPER DPR were associated with higher lesion conspicuity and contrast-related metrics, at the expense of increased noise. Images reconstructed with a 256 × 256 matrix showed lower background variability than those reconstructed with a 512 × 512 matrix. In the clinical study, these reconstruction settings resulted in higher SUVmax and tumor-to-background ratios for IMLN metastases, and visual scores for diagnostic confidence were higher for HYPER Iterative (RS = 0.7-0.91) and uAI HYPER DPR (Str = 2-4) than for OSEM.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：