EJNMMI Physics最新文献

Objectives: The present study aimed to investigate the influence of the deep progressive learning reconstruction (DPR) algorithm on the ¹⁸F-FDG PET image quality and quantitative parameters.

Methods: In this retrospective study, data were collected from 55 healthy individuals and 184 patients with primary malignant pulmonary tumors who underwent ¹⁸F-FDG PET/CT examinations. PET data were reconstructed using the ordered subset expectation maximization (OSEM) and DPR algorithms. The influence of DPR algorithm on quantitative parameters was explored, including the SUV_max, SUV_mean, standard deviation of SUV (SUV_SD), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and tumor-to-background uptake ratio (TBR). Finally, the differences in image quality parameters, including signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), between the two reconstruction algorithms were evaluated.

Results: DPR algorithm significantly reduced the SUV_max and SUV_SD of background tissues (all, P < 0.001) compared to OSEM algorithm, while no statistical difference was observed in SUV_mean between the two algorithms (all, P > 0.05). DPR algorithm notably increased the SUV_max, SUV_mean, and TBR of lesions (all, P < 0.001) and reduced MTV (P = 0.005), with minimal differences in TLG noted between the reconstruction algorithms (P < 0.001). The percentage differences in SUV_max (P = 0.001), SUV_mean (P = 0.005), and TBR (P = 0.001) between the two algorithms were significantly higher in solid nodules than in pure ground glass nodules (pGGNs). The ΔCNR between solid nodules (P = 0.031) and mixed ground glass nodules (P = 0.020) was greater than that between pGGNs. SNR and CNR obtained using the DPR algorithm were markedly improved compared to those determined using the OSEM algorithm (all, P < 0.001).

Conclusion: Under identical acquisition conditions, the DPR algorithm enhanced the accuracy of quantitative parameters in pulmonary lesions and potentially improved lesion detectability. The DPR algorithm increased image SNR and CNR compared to those obtained using the OSEM algorithm, significantly optimizing overall image quality. This advancement facilitated precise clinical diagnosis, underpinning its potential to significantly contribute to the field of medical imaging.

Purpose: This study aimed to determine a mathematical model for accurately calculating time-integrated activities (TIAs) of target tissue in ¹³¹I therapy for benign thyroid disease using the population-based model selection and non-linear mixed-effects (PBMS-NLME) method.

Methods: Biokinetic data of ¹³¹I in target tissue were collected from seventy-three patients at 2, 6, 24, 48, and 96 (N = 53) or 120 (N = 20) h after oral capsule administration with 1 MBq ¹³¹I. Based on the Akaike weight, the best sum-of-exponential function (SOEF) describing the biokinetic data was selected using PBMS-NLME modelling. Nine SOEF with three to six parameters (including the function from the European Association of Nuclear Medicine Standard Operational Procedure (EANM SOP)) were used. The fittings were repeated 1000 times with different starting values of the SOE parameters to find the optimal fit. Akaike weight was used to identify the performance of the best model from PBMS-NLME and the EANM SOP SOE function with individual fitting.

Results: Based on the PBMS-NLME analysis, the SOEF $\frac{{\lambda }_1}{{\lambda }_2+{\lambda }_1-{\lambda }_3}\left(e^{-\left({\lambda }_3+{\lambda }_{\mathrm{phys}}\right)t}-e^{-\left({\lambda }_1+{\lambda }_2+{\lambda }_{\mathrm{phys}}\right)t}\right)+a_1{e}^{-\left({\lambda }_1+{\lambda }_2+{\lambda }_{\mathrm{phys}}\right)t}$ was selected as the function most supported by the data. The Akaike weight of the best function was approximately 100%. The best SOEF from the PBMS-NLME approach shows a better performance in describing the biokinetic data of ¹³¹I in the thyroid gland than the function from the EANM SOP with individual fitting, based on the Akaike weight.

Conclusions: The best mathematical model from the PBMS-NLME approach has one more free parameter than the EANM SOP function, which could lead to more accurate TIAs.

Background: Patients' diagnosis, treatment and follow-up increasingly rely on multimodality imaging. One of the main limitations for the optimal implementation of hybrid systems in clinical practice is the time and expertise required for applying standardized protocols for equipment quality assurance (QA). Experimental phantoms are commonly used for this purpose, but they are often limited to a single modality and single quality parameter, lacking automated analysis capabilities. In this study, we developed a multimodal 3D-printed phantom and software for QA in positron emission tomography (PET) hybrid systems, with computed tomography (CT) or magnetic resonance (MR), by assessing signal, spatial resolution, radiomic features, co-registration and geometric distortions.

Results: Phantom models and Python software for the proposed QA are available to download, and a user-friendly plugin compatible with the open-source 3D-Slicer software has been developed. The QA viability was proved by characterizing a Philips-Gemini-TF64-PET/CT in terms of signal response (mean, µ), intrinsic variability for three consecutive measurements (daily variation coefficient, CoV_d) and reproducibility over time (variation coefficient across 5 months, CoV_m). For this system, averaged recovery coefficient for activity concentration was µ = 0.90 ± 0.08 (CoV_d = 0.6%, CoV_m = 9%) in volumes ranging from 7 to 42 ml. CT calibration-curve averaged over time was $\text{HU}=(951\pm 12)\times \text{density}-(944\pm 15)$ with variability of slope and y-intercept of (CoV_d = 0.4%, CoV_m = 1.2%) and (CoV_d = 0.4%, CoV_m = 1.6%), respectively. Radiomics reproducibility resulted in (CoV_d = 18%, CoV_m = 30%) for PET and (CoV_d = 15%, CoV_m = 22%) for CT. Co-registration was assessed by Dice-Similarity-Coefficient (DSC) along 37.8 cm in superior-inferior (z) direction (well registered if DSC ≥ 0.91 and Δz ≤ 2 mm), resulting in 3/7 days well co-registered. Applicability to other scanners was additionally proved with Philips-Vereos-PET/CT (V), Siemens-Biograph-Vison-600-PET/CT (S) and GE-SIGNA-PET/MR (G). PET concentration accuracy was (µ = 0.86, CoV_d = 0.3%) for V, (µ = 0.87, CoV_d = 0.8%) for S, and (µ = 1.10, CoV_d = 0.34%) for G. MR(T2) was well co-registered with PET in 3/4 cases, did not show significant distortion within a transaxial diameter of 27.8 cm and along 37 cm in z, and its radiomic variability was CoV_d = 13%.

Conclusions: Open-source QA protocol for PET hybrid systems has been presented and its general applicability has been proved. This package facilitates simultaneously simple and semi-a

Background: Medical imaging data frequently encounter image-generation heterogeneity and class imbalance properties, challenging strong generalized predictive performances with data-driven machine-learning methods. The purpose of this study was to investigate the impact of harmonization and oversampling methods on multi-center imbalanced datasets, with specific application to PET-based radiomics modeling for histologic subtype prediction in non-small cell lung cancer (NSCLC).

Methods: The retrospective study included 245 patients with adenocarcinoma (ADC) and 78 patients with squamous cell carcinoma (SCC) from 4 centers. Utilizing 1502 radiomics features per patient, we trained, validated, and tested 4 machine-learning classifiers, to investigate the effect of no harmonization (NoH) or 4 feature harmonization methods, paired with no oversampling (NoO) or 5 oversampling methods on subtype prediction. Model performance was evaluated using the average area under the ROC curve (AUROC) and G-mean via 5 times 5-fold cross-validations. Statistical comparisons of the combined models against baseline (NoH + NoO) were performed for each fold of cross-validation using the DeLong test.

Results: The number of cross-combinations with both AUROC and G-mean outperforming baseline in validation and testing was 15, 4, 2, and 7 (out of 29) for random forest (RF), linear discriminant analysis (LDA), logistic regression (LR), and support vector machine (SVM), respectively. ComBat harmonization combined with oversampling (SMOTE) via RF yielded better performance than baseline (AUROC and G-mean of validation: 0.725 vs. 0.608 and 0.625 vs. 0.398; testing: 0.637 vs. 0.567 and 0.506 vs. 0.287), though statistical significances were not observed.

Conclusions: Applying harmonization and oversampling methods in multi-center imbalanced datasets can improve NSCLC-subtype prediction, but the effect varies widely across classifiers. We have created open-source comparisons of harmonization and oversampling on different classifiers for comprehensive evaluations in different studies.

Background: Magnetic resonance (MR) images have been applied in diagnostic and therapeutic nuclear medicine to improve the visualization and characterization of soft tissues and tumors. However, the physical density (ρ) and elemental composition of human tissues required for dosimetric calculation cannot be directly converted from MR images, obstructing MR-based personalized internal dosimetry. In this study, we proposed a method to derive physical densities from Dixon MR images for voxel-based internal dose calculation.

Methods: The proposed method defined human tissues as composed of four basic tissues. The physical densities of the human tissues were calculated using the standard tissue composition of the basic tissues and the volume fraction maps calculated from Dixon images. The derived ρ map was applied to calculate the whole-body internal dosimetry using a multiple voxel S-value (MSV) approach. The accuracy of the proposed method in deriving ρ and calculating the internal dose of ¹⁸F-FDG PET imaging was evaluated by comparing with those obtained from computed tomography (CT) images of the same patient and was compared with those obtained using generative adversarial networks (GANs).

Results: The proposed method was superior to the GANs in deriving ρ from Dixon MR images and the following internal dose calculation. On average of a validation set, the mean absolute percent errors (MAPEs) of the whole-body ρ derivation and internal dose calculation using the proposed method were 14.28 ± 11.11% and 3.31 ± 0.69%, respectively. The MAPEs were respectively reduced to 5.97 ± 2.51 and 2.75 ± 0.69% after excluding the intestinal gas with different locations in the Dixon MR and CT images.

Conclusions: The proposed method could be applied for accurate and efficient personalized internal dosimetry evaluation in MR-integrated nuclear medicine clinical applications.

Background: A deep progressive learning method for PET image reconstruction named deep progressive reconstruction (DPR) method was developed and presented in previous works. It has been shown in previous study that the DPR with one-third duration can maintain the image quality as OSEM with standard dose (3.7 MBq/kg). Subsequent studies have shown we can reduce the administered activity of ¹⁸F-FDG by up to 2/3 in a real-world deployment with DPR. The aim of this study is to assess the impact of the use of DPR on Deauville score (DS) and clinical interpretation of PET/CT in patients with lymphoma.

Methods: A total of 87 lymphoma patients (age, 45.1 ± 14.9 years) who underwent ¹⁸F-FDG PET imaging for during or post-treatment follow-up from November 2020 to February 2024 were prospectively enrolled. The patients were randomly assigned to two groups, including the 1/3 standard dose group and the standard dose group. Forty-four patients were injected with 1/3 standard dose (1.23 MBq/kg) and scanned for 6 min per bed and were reconstructed: ordered-subsets expectation maximization (OSEM) with 6 min per bed (OSEM_6 min_1/3), OSEM_2 min_1/3 and DPR_2 min_1/3. Forty-three patients were scanned according to the standard protocol (3.7 MBq/kg) and were reconstructed: OSEM with 2 min per bed (OSEM_2 min_full), OSEM_40 s_full and DPR_40 s_full. Additionally, the conventional 5-point scale measurement analysis was performed and DS for lymphoma were determined in different groups. Wilcoxon signed-rank test was used to compare the mean values of liver SUVmax and mediastinal blood pool (MBP) SUVmax in each group. Likert scale and DS were evaluated using Wilcoxon signed rank test.

Results: The patients with OSEM_6 min_1/3 and DPR_2 min_1/3 showed good image quality with 5(5,5) and 5(4,5) of Likert scoring, as well as the patients with OSEM_2 min_full and DPR_40 s_full. No significant difference was found between the OSEM_6 min_1/3 and DPR_2 min_1/3 groups in terms of liver SUVmax and MBP SUVmax (P = 0.452 and 0.430), as well as the patients with OSEM_2 min_full and DPR_40 s_full (P = 0.105 and 0.638). No significant difference was found between the OSEM_6 min_1/3 and DPR_2 min_1/3 groups in terms of lesion SUVmax (P = 0.080). There was a significant differences in lesion SUVmax between OSEM-2 min_full with DPR-40 s_full (P = 0.027). The DS results were consistent (100%) between OSEM-6 min_1/3 with DPR_2 min_1/3, and between OSEM-2 min_full with DPR-40 s_full, respectively.

Conclusions: DPR reconstruction demonstrated feasibility in reducing PET injection dose or scanning time, while ensuring the preservation of image quality and DS for during or post-treatment follow-up patients with lymphoma.

Background: Diagnostic imaging is a dynamic medical field. In nuclear medicine, advancements introduce new procedures utilising innovative radiopharmaceuticals. These developments may influence supply requirements and exposure levels for the patient population. Surveying the frequency of procedures, types of pharmaceuticals, and administered activities provides valuable insights into utilisation trends and radionuclide demand. This knowledge also guides the prioritisation of radiation protection efforts at national and local levels. In Europe, radiation dose assessments for medical exposures are mandatory according to the directive´s requirements.

Methods: This study evaluated the utilisation of diagnostic nuclear medicine procedures in Sweden over 15 years (2008-2023), focusing on procedure frequency, effective dose, and collective effective dose. Comprehensive data from all Swedish clinics performing nuclear medicine were analysed, incorporating information on radiopharmaceuticals and administered activities. The method suggested by the UNSCEAR, which includes so-called essential procedures, was used for comparison. The study also investigated some frequent procedures in more detail.

Results: The study identifies noteworthy trends, including a threefold increase in the number of clinics offering Positron Emission Tomography (PET) procedures and a significant rise in PET usage. PET procedures constituted over 50% of the collective effective dose for adults in 2023. Despite this, Gamma Camera (GC) procedures still dominate in frequency but exhibit a steady decline. Procedures using ^99mTc and ¹⁸F accounted for 93% of procedures in 2023. The collective effective dose rose 22% over the study period, with PET procedures driving this increase. PET procedures increasing role became evident by the increased contribution to the total collective dose from 15 to 52%. The UNSCEAR methodology captured 67% of the total frequency and underestimated the collective effective dose by 16%. Administered activity remained stable for the selected procedures and showed low variation between clinics.

Conclusions: PET procedures are increasing in scope and now constitute the largest contribution to radiation dose, and in-house production of PET radiopharmaceuticals is available in around 40% of clinics. The number of radionuclides decreased over the study period, and GC procedures declined. In general, the amount of administered activity remained stable over the period for the procedures studied. Accurately assessing utilisation and exposure trends requires extensive data, and the methodology used affects the result significantly.

Purpose: Synaptic vesicle glycoprotein 2 A (SV2A) in human brains is an important biomarker of synaptic loss associated with several neurological disorders. However, SV2A tracers, such as [¹¹C]UCB-J, are less available in practice due to constrains such as cost, radiation exposure and onsite cyclotron. We therefore aim to generate synthetic [¹¹C]UCB-J PET images based on MRI in this study.

Methods: We implemented a convolution-based 3D encoder-decoder to predict [¹¹C]UCB-J SV2A PET images. A total of 160 participants who underwent both MRI and [¹¹C]UCB-J PET imaging, including individuals with schizophrenia, cannabis use disorder, Alzheimer's disease, were used in this study. The model was trained on pairs of T1-weighted MRI and [¹¹C]UCB-J distribution volume ratio images, and tested through a 10-fold cross-validation process. The image translation accuracy was evaluated based on the mean squared error, structural similarity index, percentage bias and Pearson's correlation coefficient between the ground truth and the predicted images. Additionally, we assessed the prediction accuracy of selected regions of interest (ROIs) crucial for brain disorders to evaluate our results.

Results: The generated SV2A PET images are visually similar to the ground truth in terms of contrast and tracer distribution, quantitatively with low bias (< 2%) and high similarity (> 0.9). Across all diagnostic categories and ROIs, including the hippocampus, frontal, occipital, parietal, and temporal regions, the synthetic SV2A PET images exhibit an average bias of less than 5% compared to the ground truth. The model also demonstrates a capacity for noise reduction, producing images of higher quality compared to the low-dose scans.

Conclusion: We conclude that it is feasible to generate robust SV2A PET images with promising accuracy from MRI via a data-driven approach.

Background: The high sensitivity and axial coverage of large axial field of view (LAFOV) PET scanners have an unmet potential for total-body PET research. Despite these technological advances, inherent challenges to PET scans such as patient motion persist. To provide simulation-derived ground truth information, we developed a digital replica of the Biograph Vision Quadra LAFOV PET/CT scanner closely mimicking real event processing and image reconstruction.

Material and methods: The framework uses a GATE model in combination with vendor-specific software prototypes for event processing and image reconstruction (e7 tools, Siemens Healthineers). The framework was validated against experimental measurements following the NEMA NU-2 2018 standard. In addition, patient-like simulations were performed with the XCAT phantom, including respiratory motion and modeled lesions of 5, 10, 20 mm size, to assess the impact of motion artefacts on PET images using a motion-free reference.

Results: The simulation framework demonstrated high accuracy in replicating scanner performance in terms of image quality, contrast recovery (37 mm sphere: 86.5% and 85.5%; 28 mm: 82.6% and 82.4%; 22 mm: 78.8% and 77.7%; 17 mm: 74.9% and 74.6%; 13 mm: 67.0% and 67.9%; 10 mm: 55.5% and 64.3%), image noise (CV of 7.5% and 7.7%) and sensitivity (174.6 cps/kBq and 175.3 cps/kBq) for the simulation and experimental data, respectively. High agreement was found for the spatial resolution with a difference of 0.4 ± 0.3 mm and the NECR aligned well with a maximum deviation of 9%, particularly in the clinical activity range below 10 kBq/mL. Motion induced artefacts resulted in a quantification error at lesion level between - 12.3% and - 45.1%.

Conclusion: The experimentally validated digital twin of the Biograph Vision Quadra facilitates detailed studies of realistic patient scenarios while offering unprecedented opportunities for motion correction, dosimetry, AI training, and imaging protocol optimization.