Nuclear Medicine Communications最新文献

Purpose: This study aimed to evaluate the influence of silicon photomultiplier PET (SiPM-PET) and the Bayesian penalized-likelihood (BPL) reconstruction algorithm on the measurement of metabolic tumor volume (MTV), in comparison with conventional photomultiplier tube PET (PMT-PET).

Materials and methods: Six phantoms of varying shapes and volumes (1.2 to 20 ml) were prepared using an agar-based fluorodeoxyglucose F-18 mixture with a background-to-lesion activity ratio of 1 : 4 and placed in a 5-L water tank. Each phantom was scanned using PMT-PET and SiPM-PET systems. PMT-PET images were reconstructed with conventional ordered-subset expectation maximization (OSEM), whereas SiPM-PET data were reconstructed using both OSEM and BPL. MTV was calculated using relative threshold-based segmentation [30, 40, 42, and 50% of maximum standardized uptake value (SUVmax)] and a gradient-based method with weight coefficients of 0.3, 0.4, 0.5, and 0.6. Measured MTVs were compared with true phantom volumes to assess accuracy.

Results: SiPM-PET provided a more accurate geometric representation of phantom contours than PMT-PET. Across all conditions, a 40% SUVmax threshold and a weight coefficient of 0.4 yielded MTVs closest to the true volumes. When SiPM-PET was combined with BPL, gradient-based segmentation produced MTVs with the smallest deviation from the known volumes, particularly for irregularly shaped phantoms.

Conclusion: SiPM-PET using BPL facilitated clearer delineation of phantom boundaries compared with PMT-PET and SiPM-PET with OSEM. Under the present experimental conditions, MTVs obtained from SiPM-PET with BPL, especially when using the gradient-based method, showed improved agreement with true phantom volumes, suggesting potential advantages for volumetric assessment in oncologic PET imaging.

Objective: The objective of this study is to evaluate the combined prognostic values of 18F-fluorodeoxyglucose (18F-FDG) PET and computed tomography (CT)-derived entropy-based heterogeneity features from hybrid PET/CT scanner using machine learning in patients with lung adenocarcinoma undergoing curative surgery.

Methods: Presurgical 18F-FDG PET/CT from 131 patients with lung adenocarcinoma were divided into training (n = 92) and temporal validation (n = 39) cohorts. In the training cohort, we integrated entropy-based heterogeneity features from 18F-FDG PET/CT for disease-free survival (DFS) prediction using machine learning approach. The predictive value of clinical variables and 18F-FDG PET/CT-based machine learning for DFS was examined using Cox regression analyses, and independent prognosticators were used to develop the survival prediction model. The model was then tested in the temporal validation cohort.

Results: In the training cohort, 18F-FDG PET/CT-based machine learning, female sex, and pN status independently predicted DFS. The model, incorporating these predictors significantly predicted DFS in the training (hazard ratio = 1.483, P < 0.001) and validation cohorts (hazard ratio = 1.753, P < 0.001). This model outperformed traditional staging system in both cohorts (c-indices = 0.717 vs. 0.621 in training; and 0.728 vs. 0.644 in validation). The model also predicted overall survival in both cohorts (hazard ratio = 1.370, P < 0.001 in training; hazard ratio = 1.574, P = 0.017 in validation).

Conclusion: Our preliminary results suggest that integrating prognostic values from 18F-FDG PET and CT-based heterogeneity features with clinical prognosticators is feasible and may support personalized treatment strategies for patients with resectable lung adenocarcinoma.

Objectives: The objective of this study is to evaluate the efficacy of multi-parametric MRI (mpMRI) and prostate-specific membrane antigen (PSMA) PET-computed tomography (CT) in preoperative loco-regional assessment of prostate carcinoma, correlating findings with histopathology.

Materials and methods: This prospective observational study enrolled 44 men (mean age 67.5 years) with suspected localized prostate cancer. All participants underwent mpMRI and PSMA PET-CT scans prior to histopathological confirmation via transrectal ultrasound-guided biopsies or radical prostatectomy. Imaging results were analyzed for sensitivity, specificity, and agreement with histopathology using Cohen's Kappa statistic. Key parameters, such as Prostate Imaging Reporting and Data System (PI-RADS) scores (mpMRI) and maximum standardized uptake (SUVmax) values (PSMA PET-CT), were correlated with Gleason scores.

Results: MpMRI exhibited a sensitivity of 95% and specificity of 50%, while PSMA PET-CT achieved 96.3% sensitivity and 57.1% specificity. MpMRI showed strong agreement with histopathology for tumor site localization (Kappa = 0.760), surpassing PSMA PET-CT (Kappa = 0.651). PSMA PET-CT identified metastases in 27.2% of cases, while mpMRI detected extra-prostatic extension in 50%. Higher PI-RADS scores and SUVmax values were associated with increased Gleason scores, indicating aggressive disease.

Conclusion: Both mpMRI and PSMA PET-CT offer high sensitivity for prostate cancer detection. MpMRI is superior for local staging, while PSMA PET-CT excels in identifying distant metastases. Their combined application enhances diagnostic accuracy and supports improved preoperative risk stratification in prostate carcinoma management.

Background: Graves' disease (GD), a leading cause of hyperthyroidism, exhibits heterogeneous responses to iodine-131 therapy, underscoring the need for accurate predictive tools. While pertechnetate (99mTcO4-) thyroid scintigraphy provides essential functional insights, its clinical utility is restricted by limited resolution and the subjective interpretation of results. Deep learning emerges as a promising approach for extracting prognostic features from imaging data.

Objective: This research seeks to construct a radiomic model employing deep learning methodologies, utilizing thyroid scintigraphy, to predict the efficacy of initial iodine-131 therapy in patients diagnosed with GD.

Methods: This retrospective study involved 121 patients diagnosed with GD who underwent pretreatment thyroid scintigraphy. A ResNet50 convolutional neural network was implemented, trained end-to-end on standardized thyroid scintigraphy data to classify treatment outcomes as either cured or not cured. Gradient-weighted class activation mapping (Grad-CAM) was employed to identify region-specific predictive features. Radiomic features extracted from deep learning were further analyzed using least absolute shrinkage and selection operator regression, with a random forest (RF) model serving as the final predictive tool. Performance metrics, including the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity, were evaluated on independent training and validation datasets, using an 8 : 2 split.

Results: The ResNet50 model yielded AUC values of 0.828 [95% confidence interval (CI): 0.740-0.915] and 0.806 (95% CI: 0.656-0.956) for the training and validation datasets, respectively. A set of 16 critical radiomic features, extracted from the global pooling layer of ResNet50, enabled the construction of an RF classifier that exhibited enhanced generalization performance, as evidenced by a test AUC of 0.924. Decision curve analysis substantiated the model's clinical utility across a broad spectrum of threshold probabilities (10-90%). Grad-CAM visualizations revealed that the model predominantly concentrated on regions of thyroid parenchyma associated with treatment-responsive tissue.

Conclusion: This study introduces a pioneering deep learning framework that employs thyroid scintigraphy radiomics to predict iodine-131 therapeutic outcomes in GD. By combining ResNet50 for feature extraction with interpretable Grad-CAM localization, this approach advances personalized nuclear medicine strategies and mitigates the 'black box' limitation typically associated with artificial intelligence models. These findings require validation across multicenter cohorts to refine and optimize precision treatment protocols.

Objective: To evaluate the relationship between posttreatment 18 F-fluorodeoxyglucose ( 18 F-FDG) PET/computed tomography (CT) restaging with metabolic parameters and prognosis of uterine cervical cancer.

Methods: A total of 151 patients with cervical cancer who underwent 18 F-FDG PET/CT examination were retrospectively analyzed. The correlation between PET/CT restaging, maximum standard uptake value, whole body metabolic tumor volume (wbMTV), whole body total lesion glycolysis (wbTLG), related clinical factors, and prognosis was analyzed. Kaplan-Meier survival analysis was used to perform univariate analysis on clinical parameters and imaging parameters, and Cox's proportional hazards regression model was used to perform multifactor analysis to explore the related factors affecting progression-free survival (PFS) and overall survival (OS) of patients with cervical cancer.

Results: With a median follow-up time of 24 months, the median OS of 151 cervical cancer patients was 43 months, and the 5-year survival rate was 50%; the median PFS was 33 months, and the 5-year PFS rate was 29%. Both univariate and multivariate analyses showed that PET/CT restaging and squamous cell carcinoma antigen (SCC-Ag) were significant prognostic factors for OS and PFS. Moreover, patients with downstaged PET/CT restaging showed significantly better 5-year OS and 5-year PFS rates than those no downstaging (87.0 vs. 24.6%; P  < 0.001 for OS; 50.7 vs. 18.5%; P  < 0.001 for PFS). Among patients with positive PET/CT findings, wbTLG and wbMTV were identified as independent prognostic factors for OS and PFS, respectively.

Conclusion: 18 F-FDG PET/CT restaging, serum SCC-Ag, wbMTV, and wbTLG are established as prognostic biomarkers in recurrent cervical cancer. Posttreatment PET/CT represents a sensitive and accurate imaging technique for predicting patient outcomes.

Objectives: In primary hyperparathyroidism, preoperative localisation of a parathyroid adenoma enables focussed excision rather than full neck exploration, with reduced complication rates, recovery time and cost. In this study, we sought to improve parathyroid adenoma localisation on 99m Tc-sestamibi so more patients can benefit from focussed excision.

Methods: We performed a baseline audit of our 99m Tc-sestamibi service, followed by three sequential changes: cycle 1: acquisition of new single-photon emission computed tomography/computed tomography (SPECT/CT) scanners, cycle 2: optimisation of image acquisition and reconstruction, cycle 3: addition of arterial phase iodinated contrast to SPECT/CT. We classified reporter confidence using a 5-point Likert scale. We correlated adenoma location on imaging with 'gold standard' location based on surgical and pathology results. We calculated the sensitivity, specificity, and receiver operator characteristics following each change cycle.

Results: Results comparing each cycle to the baseline audit: Baseline: 44% confidently located, 31% failed to identify an adenoma. Sensitivity: 69.8%, specificity: 95.7%, area under the curve (AUC): 0.92. Cycle 1: 54% confidently located, 24% failed to identify an adenoma. Sensitivity: 77.6%, specificity: 93.3%, AUC: 0.95. Cycle 2: 48% confidently located, 18% failed to identify an adenoma. Sensitivity: 84.3%, specificity: 95.9%, AUC: 0.98. Cycle 3: 67% confidently located, 6% failed to identify an adenoma. Sensitivity: 94.7%, specificity: 90%, AUC: 0.99 ( P  < 0.001).

Conclusion: Our study demonstrated that sequential improvements in 99m Tc-sestamibi SPECT/CT significantly improved test performance by reducing the number of localisation failures, increasing test sensitivity, and reducing the number of false negatives. As a result, more patients can benefit from focussed excision parathyroidectomy.

Tumor thrombus (TT) represents the intravascular extension of the tumor, most commonly seen in renal cell carcinoma (RCC) and hepatocellular carcinoma (HCC). It is clinically significant with direct implications on staging, prognosis, and therapeutic decision-making. TT is most commonly noted in RCC and HCC, but can also occur in several other malignancies, such as adrenal cortical carcinoma, colorectal carcinoma, pancreatic neuroendocrine tumors, thyroid carcinoma, and testicular carcinoma. Differentiating TT and bland thrombus is crucial. Surgical resection offers survival benefit in select cases, whereas systemic therapies, including immunotherapy and targeted agents, are transforming the management of RCC and HCC with vascular invasion. TT is a marker of aggressive tumor biology that significantly worsens the prognosis. This review highlights current knowledge about the pathogenesis, imaging features, prognostic impact, and therapeutic approaches to TT, with emphasis on the evolving role of nuclear medicine, and provides a narrative review of peer-reviewed literature and available guidelines.

Background: [18F]fluoro-2-deoxy-D-glucose ([18F]FDG)-PET/MRI is a 'one-stop-shop' combination that enriches lesion characterization, locoregional and distant staging in breast cancer patients. The combination consists of a breast (bpPET/MRI) and whole-body (wbPET/MRI) within a single protocol. Despite acceptable test-retest reproducibility of standardized uptake value (SUV) in PET examinations being 10%, limited knowledge concerning agreement between metabolic parameters from each protocol (bpPET/MRI and wbPET/MRI) is available.

Objective: Feasibility study to verify whether differences of metabolic parameters in breast tumor, on prone bpPET/MRI acquired between 30 and 55 min postinjection, and compared to wbPET/MRI at 60 min, remained within the 10% margin acceptable in clinical practice.

Methods: Breast cancer patients underwent [18F]FDG-PET/MRI: bpPET/MRI and wbPET/MRI. Maximum SUV (SUVmax) and mean SUV (SUVmean) measurements of the index tumor, contralateral breast, liver, and blood pool were obtained from bpPET/MRI images acquired every 5 min between 30 and 55 min. These values were compared with wbPET/MRI SUVs at 60 min (reference).

Results: Twenty patients were included. The majority of PET/MRI exams remained within the 10% margin at all timepoints. At 45 and 50 min the mean percentage difference of SUVmax and SUVmean compared to 60 min was less than 1.3-3.6%. Other SUV measurements remained within 10% margin in majority of cases, while bloodpool SUVs did differ.

Conclusion: Metabolic parameters for local staging in the index tumor on bpPET/MRI may be considered comparable to the metabolic parameters retrieved from wbPET/MRI in breast cancer patients. These findings reinforce the potential of a breast-dedicated protocol integrated in the whole-body [18F]FDG PET/MRI in breast cancer patients.

Rationale and objectives: Best practices for quantifying longitudinal changes in 99mTc-sestamibi uptake with molecular breast imaging (MBI) have not been established. The objective of this work is to quantify the variability of tumor-to-normal tissue 99mTc-sestamibi uptake measurements in patient data and the resulting impact on categorizing uptake changes throughout neoadjuvant systemic therapy (NST).

Materials and methods: In 60 patients with locoregional breast cancer undergoing NST, tumor-to-normal-tissue ratios of average contour counts () and activity concentrations () were calculated from physician contours on craniocaudal and mediolateral-oblique MBI images acquired at pre-NST, mid-NST, and post-NST with a commercial dual-headed system. Measurement variability was quantified in pre-NST images by using different contours and detector views. This analysis was then used to define patient-specific thresholds above which calculated and values at mid-NST and post-NST imaging had 'changed' from pre-NST values.

Results: Pre-NST values ranged from 1-5 to 1-100 and in general decreased by 10-30% at mid-NST and by 50-70% at post-NST. uncertainty was significantly higher than uncertainty (53% vs. 35% median value; paired t-test; P < 0.007), driven primarily by normal-tissue contour placement within the heterogeneous signal in normal breast parenchyma. In paired comparisons, a significantly higher number of cases 'changed' when the metric was derived from a single image instead of averaged across images (McNemar's test, P < 0.0125).

Conclusion: By ensuring consistent acquisition conditions and standardizing contour definitions, robust metrics of tumor and normal-tissue uptake can be extracted from heterogenous clinical MBI data for longitudinal quantitative evaluations.