EJNMMI Physics最新文献

{"title":"Multi-modal segment anything model (mmSAM) for tumor segmentation in multi-tracer oncologic PET/CT.","authors":"Zhijie Fang, Yibin Liu, Tao Tan, Feng Wang, Greta S P Mok","doi":"10.1186/s40658-026-00887-z","DOIUrl":"10.1186/s40658-026-00887-z","url":null,"abstract":"<p><strong>Purpose: </strong>Lesion segmentation is important in targeted radionuclide therapy dosimetry. In this study, we proposed a novel 2D multi-modal Segment Anything Model (mmSAM) for lesion segmentation in whole-body multi-tracer PET/CT images.</p><p><strong>Methods: </strong>The AutoPET 2024 dataset, including ¹⁸F-PSMA (369 subjects), ¹⁸F-FDG (170 subjects) and ⁶⁸Ga-PSMA (170 subjects) PET/CT images and their corresponding tumor masks, was used in this study. The ¹⁸F-PSMA PET/CT dataset was used as the primary dataset and was divided into 233: 36: 100 for training, validation and testing the mmSAM, using both PET and CT input. The transferability of the primary model was evaluated on 100 patients on other 2 datasets, without and with fine-tuning using 70 cases of ¹⁸F-FDG and ⁶⁸Ga-PSMA respectively. Standard single modal 2D SAM with only PET input, 3D nnUNet with two-channel PET/CT input and the thresholding-based method were also implemented for comparison. Mean Dice, the 95th percentile Hausdorff distance (mean HD95), mean standardized uptake value (mean |SUV_mean| error), metabolic tumor volume (mean |MTV| error), true positive rate (TPR), positive predictive value (PPV), and false discovery rate (FDR) were computed for all tumors. Statistical significance among different segmentation methods was evaluated using the Wilcoxon test.</p><p><strong>Results: </strong>mmSAM achieved the best performance as compared to other segmentation methods for the primary ¹⁸F-PSMA dataset (mean Dice/HD95/|SUV_mean| error/|MTV| error/TPR/PPV/FDR = 0.76/1.92 mm/5.10%/14.60%/100%/96.51%/3.49%, all p < 0.05), without fine-tuning (mean Dice/HD95/|SUV_mean| error/|MTV| error/TPR/PPV/FDR = 0.61/2.28 mm/ 15.83%/27.71%/100%/78.91%/21.09% for ¹⁸F-FDG; 0.77/1.33 mm/5.55%/17.96%/100% /97.78%/2.22% for ⁶⁸Ga-PSMA, p < 0.05), as well as with fine-tuning (mean Dice/HD95/|SUV_mean| error/|MTV| error/TPR/PPV/FDR = 0.65/1.81 mm/6.62%/13.93%/100%/ 80.50%/19.50% for ¹⁸F-FDG; 0.81/1.15 mm/4.62%/14.30%/100%/99.30%/0.70% for ⁶⁸Ga-PSMA, p < 0.05) on the cross-tracer datasets.</p><p><strong>Conclusion: </strong>The proposed mmSAM is promising for lesion segmentation in multi-tracer oncologic PET/CT images. Fine-tuning significantly enhances segmentation accuracy for cross-tracer studies.</p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13346370/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147962622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantifying capillary transit time distribution: a novel tracer kinetic model accounting for leakiness, back diffusion, and perfusion in MRI and PET imaging-the outflow model.","authors":"Henrik B W Larsson, Stig P Cramer, Edis D Tireli, Tanne S W Larsson, Ulrich Lindberg, Mark B Vestergaard","doi":"10.1186/s40658-026-00886-0","DOIUrl":"10.1186/s40658-026-00886-0","url":null,"abstract":"<p><strong>Background: </strong>Brain blood perfusion is typically characterized through regional cerebral blood flow metrics. However, recent research demonstrates that the distribution of blood transit times through the capillary network also plays a significant role in physiological processes. Here, we introduce a novel kinetic model, called the Outflow model, to quantify this distribution and calculate capillary transit time heterogeneity. The model is tailored to account for transport across the blood‒brain barrier and tissue-binding and is applicable for both for Magnetic Resonance Imaging and Positron emission Tomography scenarios. The model was employed across three methodologies: (1) contrast-enhanced T1-weighted Magnetic Resonance Imaging; (2) dynamic long axial field-of-view Positron Emission Tomography imaging using the radiotracers O-(2-[¹⁸F]fluoroethyl)-L-tyrosine and (3) 2-deoxy-2-[¹⁸F]fluoroglucose. The temporal resolution was optimized (1-2 s) to capture the bolus passages. Tissue curves were derived from the putamen, thalamus, frontal white matter, and regions of malignancy for comparative analysis. For the PET data the Outflow model was compared with a conventional 3-compartment model for validation.</p><p><strong>Results: </strong>For Magnetic Resonance Imaging, the capillary transit time heterogeneity ranged from 0.5 to 1 s, the extraction fraction ranged from 0.03 to 0.07%, and the unidirectional influx constant ranged from 0.01 to 0.2 ml/min/100 ml. For O-(2-[¹⁸F]fluoroethyl)-L-tyrosine, the capillary transit time heterogeneity ranged from 1.5 to 3 s, with an extraction fraction of 5.4% to 7.8% and the unidirectional influx constant ranging from 1.6 to 3.2 ml/min/100 ml, aligning closely with the anticipated parameters. For 2-deoxy-2-[¹⁸F]fluoroglucose obtain from the basal ganglia, capillary transit time heterogeneity was around 2 s, extraction fraction 40.2%, and cerebral metabolic rate of glucose consumption was 44 µmol/100 ml/min. Both the outflow model and the 3-compartment model gave excellent fit to data and for the comparable metrics the two models gave results in reasonable agreement.</p><p><strong>Conclusion: </strong>This proof-of-concept study demonstrated that calculating capillary transit time heterogeneity in the brain via the proposed new tracer kinetic model is feasible for both MRI and PET data. The model adeptly addresses scenarios where tracers or contrast agents undergo bidirectional transport across the BBB.</p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13253907/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147926870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multiple vertebrae improves precision in image-based bone marrow absorbed dose estimation in [¹⁷⁷Lu]Lu-DOTATATE treatment.","authors":"Katja Smits, Frida Westerbergh, Jens Hemmingsson, Martijn van Essen, Johanna Svensson, Anna-Lena Theisen, Johannes Tran-Gia, Tobias Rydén, Peter Bernhardt","doi":"10.1186/s40658-026-00882-4","DOIUrl":"10.1186/s40658-026-00882-4","url":null,"abstract":"<p><strong>Background: </strong>The bone marrow is a critical dose limiting organ in patients receiving radionuclide therapy, but image-based bone marrow dosimetry is not routinely performed in clinical practice. High noise levels and the limited number of vertebral cavities used in bone marrow absorbed dose calculations make absorbed dose estimates susceptible to large uncertainties. This study explores the effects of the number of included vertebrae, partial volume effects, and reconstruction parameters on image-based bone marrow absorbed dose calculations. We included sixteen patients with advanced neuroendocrine tumors treated with [¹⁷⁷Lu]Lu-DOTATATE. Bone marrow absorbed dose estimates and their precision were analyzed based on the number of vertebrae included in the calculation. Additional evaluations were performed using both measured (Lung-Spine phantom with spherical inserts) and simulated (digital XCAT phantom) SPECT data, with noise levels matching patients' bone marrow.</p><p><strong>Results: </strong>Increasing the number of vertebrae from one to six improved precision in absorbed doses, reducing the coefficient of variation (COV) from 34 to 6.2%. Only minor differences (0.1%) in bone marrow absorbed dose were observed when adjusting reconstruction parameters from 1 subset, 60 iterations, to 12 subsets, 5 iterations. In the Lung-Spine phantom, smaller volumes (4 mL) were more sensitive to noise than larger volumes (16 mL), with COVs of 36% and 17%, respectively, at 60 updates. In the XCAT phantom, a decrease in recovery was observed in vertebrae near high-uptake regions (L5: 0.63, L1: 0.29), at 60 updates at the highest noise level.</p><p><strong>Conclusion: </strong>Including multiple lesion-free vertebrae enhances precision in image-based bone marrow absorbed dose calculations. Nevertheless, careful selection of vertebrae is important, as closely located high-uptake areas can bias absorbed dose estimates. Applying partial volume correction could mitigate spill-out effects, further improving accuracy. Additionally, the minimal differences between reconstruction parameters suggest that bone marrow absorbed dose estimates remain stable across subset configurations.</p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-05-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13161456/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147863741","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multicenter comparison of radionuclide calibrators and SPECT/CT protocols for quantitative ¹⁷⁷Lu   imaging in clinical practice.","authors":"Wies Claeys, Kristof Baete, Laurence Beels, Claire Bernard, Rachele Danieli, Yves D'Asseler, An De Crop, Michel Hesse, Victor Nuttens, Bruno Vanderlinden, Michel Koole","doi":"10.1186/s40658-026-00877-1","DOIUrl":"10.1186/s40658-026-00877-1","url":null,"abstract":"<p><strong>Purpose: </strong>Following the clinical success of [Formula: see text]-based therapies for neuroendocrine tumors and prostate cancer, accurate quantification of [Formula: see text]  using radionuclide calibrators (RNCs) and SPECT/CT is gaining importance as prerequisite for accurate treatment delivery and dosimetry. However, the lack of standardization can introduce inter-system variability, compromising multi-center clinical trials. This study aimed to assess the accuracy and variability of [Formula: see text]  measurements using RNCs and SPECT/CT across different systems and hospitals.</p><p><strong>Methods: </strong>A uniform cylindrical phantom and a NEMA phantom with hot spheres were prepared using traceable activities and imaged at 8 different hospitals using 13 SPECT/CT systems (9 conventional and 4 3D CZT). Acquisitions and reconstructions were performed using both site-specific and standardized protocols. The cylindrical phantom images were used to evaluate the system calibration and establish image calibration factors (ICFs), the NEMA images to evaluate effective resolution by calculating recovery coefficients (RCs). In parallel, two vials were measured to test RNC accuracy.</p><p><strong>Results: </strong>RNC measurements differed up to 11% between centers, while SPECT quantification of the cylindrical phantom differed up to 20%. While ICFs were consistent for systems of the same type, image quality varied strongly when using clinical protocols (36% difference in RCs in the largest sphere). Standardized reconstruction reduced variability in RCs for each system type (maximum 12% difference), regardless of acquisition protocol, but differences between system types persisted when standardizing acquisition and reconstruction (33% difference).</p><p><strong>Conclusion: </strong>Current [Formula: see text] measurement practices yield significant variability in quantification and image quality. Harmonization efforts should prioritize standardized calibration and reconstruction protocols to improve multicenter reproducibility of quantitative [Formula: see text]-SPECT/CT.</p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13168373/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147835232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancements in DOI-capable TOF-PET modules using a multi-channel high-frequency readout.","authors":"Giulia Terragni, Elena Tribbia, Joshua W Cates, Marco Pizzichemi, Johann Marton, Etiennette Auffray","doi":"10.1186/s40658-026-00870-8","DOIUrl":"10.1186/s40658-026-00870-8","url":null,"abstract":"<p><strong>Purpose: </strong>The achievement of excellent coincidence time resolution has become an essential objective in time-of-flight positron emission tomography (TOF-PET) in order to improve the signal-to-noise ratio and the quality of the reconstructed image. Moreover, to achieve a high level of spatial resolution while maintaining good sensitivity with the use of long scintillators, the impact of the gamma-ray depth of interaction (DOI) must be mitigated.</p><p><strong>Methods: </strong>To address these challenges, a matrix of sixteen 20 mm-long LYSO:Ce scintillators with depolished lateral surfaces is used, coupled to a light guide that enables light sharing within the matrix. This configuration allows for the determination of the gamma-ray DOI and the correction of its timing bias. For this light-sharing method to work, readout integration in a multi-channel scheme is required. This is achieved using a sixteen-channel low-noise, low-power, high-frequency development circuit board. This high-frequency circuit also provides a solution to enhance the time resolution by enabling a lower leading-edge threshold for the detection of the earliest photons produced with high precision.</p><p><strong>Results: </strong>The DOI-capable module achieves a DOI resolution of 2.2 ± 0.2 mm full-width-at-half-maximum (FWHM) and a CTR of 133 ± 2 ps FWHM after DOI correction on the central crystals. For comparison, a standard module consisting of sixteen polished crystals and no back light guide achieves a CTR of 130 ± 2 ps FWHM.</p><p><strong>Conclusions: </strong>Using 20 mm-long LYSO:Ce crystals and this high-frequency electronics, the DOI-capable module delivers excellent timing and energy resolution. In addition to providing time resolution comparable to that of the standard module, it also allows DOI encoding with a resolution of nearly 2 mm FWHM.</p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13323475/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147812587","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Uncertainty analysis of single-time-point organ dosimetry compared with the multi-time-point method in radioimmunotherapy.","authors":"Amit Nautiyal, Gemma Lewis, Jonathan I Gear, Kim Orchard, Matthew J Guy, Sofia Michopoulou","doi":"10.1186/s40658-026-00883-3","DOIUrl":"10.1186/s40658-026-00883-3","url":null,"abstract":"<p><strong>Background: </strong>A position paper released by the European Association of Nuclear Medicine emphasised the need for multidisciplinary engagement to establish dosimetry-based personalised treatment in Radionuclide therapy (RNT). The uncertainty analysis results often ignored in routine clinical practice should be incorporated into the dose calculations to improve the efficacy and accuracy of treatment. In this study, patients with haematological malignancies undergoing radioimmunotherapy were evaluated. Our study aimed to calculate the uncertainties associated with each parameter of the single time point (STP) dosimetry chain and compare the with multiple time points (MTP) in the bone marrow and liver results.</p><p><strong>Methods: </strong>28 patients received an intravenous injection of ¹¹¹In-besilesomab (0.17 ± 0.01GBq) for pre-therapeutic dosimetry and were subsequently treated with ⁹⁰Y-besilesomab(2.43 ± 0.53GBq). A dosimetry analysis was performed on bone marrow (BM) and liver with MTP and STP. We investigated the uncertainty in population mean effective half-life, volume, recovery coefficient, counts, measured activity, fitting parameters, time-integrated-activity, S-factors, and absorbed dose (AD) for a group of patients.</p><p><strong>Results: </strong>The mean absorbed dose per unit administered activity (DpA) to BM was 5.8 ± 1.7 mGy/MBq with MTP and 5.8 ± 1.6 mGy/MBq with STP, and to the liver was 2.9 ± 1.9 mGy/MBq with MTP and 3.1 ± 2.4 mGy/MBq with STP. The mean fractional uncertainty associated with total absorbed dose to BM was 13.18 ± 3.46% with MTP and 18.75 ± 3.22% with STP, and to liver was 5.77 ± 3.13% with MTP and 49.78 ± 25.36% with STP. A moderate positive relationship (R² = 0.7) was noted between post-injection acquisition time and AD uncertainty with STP for BM, whereas a strong positive relationship (R² = 1) was noted for the liver.</p><p><strong>Conclusion: </strong>The absorbed dose uncertainty in STP was significantly higher compared to the MTP. Incorporating the uncertainty analysis for STP dosimetry parameters in routine clinical practice is strongly recommended. The accuracy in the acquisition time, population-based half-life and fitting function for time activity curve is vital for minimising uncertainty in STP dosimetry, which is less time-consuming and easier to implement in clinical practice than MTP.</p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13319317/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147812535","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The impact of motion correction on quantitatively measured perfusion indices using cardiac-dedicated dynamic CZT-SPECT myocardial perfusion imaging.","authors":"Ajay Kumar Chaudhary, Fukai Zhao, Zekun Pang, Jianming Li","doi":"10.1186/s40658-026-00881-5","DOIUrl":"10.1186/s40658-026-00881-5","url":null,"abstract":"<p><p>BACKGROUND: Dynamic myocardial perfusion imaging (MPI) with cadmium-zinc-telluride (CZT) cardiac-dedicated single-photon emission computed tomography (SPECT) enables the quantification of myocardial blood flow (MBF), myocardial flow reserve (MFR), and non-invasive fractional flow reserve (niFFR). However, motion was a prominent challenge during dynamic acquisition that could distort these indices. This research aims to characterize the motion vector, evaluate the impact of motion correction (MC) on quantitative perfusion indices, compare MC and non-motion correction (NMC) datasets, and assess their interchangeability. METHODS: We retrospectively analyzed 149 patients who underwent clinically indicated rest/adenosine-stress dynamic MPI. All listmode SPECT and CT attenuation data were transferred to the MyoFlowQ workstation, where the listmode data were rebinned into 18 frames, yielding paired MC and NMC datasets for both phases. Frame-by-frame post-MC vectors (X, Y, Z) were extracted from the MC output to characterize motion magnitude and direction. Paired difference, association, functional bias, agreement, accuracy, and reclassification with association (MFR/niFFR) were assessed between MC and NMC. Moreover, the impact of MC on subgroups was also examined. RESULTS: Statistically significant frame-wise motion was present in all axes, with predominantly leftward (X: 57.42%/70.86%), posterior (Y: 61.14%/80.47%), and inferior (Z: 83.84%/79.61%) motion in rest/stress phases affecting indices. Overall, the rest MBF (0.01–0.02 ml/min/g) was statistically underestimated (except LAD), while stress MBF (0.05–0.08 ml/min/g), MFR (0.09–0.12), and niFFR (0.02–0.03) were statistically overestimated (except stress MBF-RCA). However, bidirectional estimations were observed for all indices with both increasing and decreasing values depending on the underlying error pattern. The association between MC and NMC values was high (> 0.9), but concordance ranged from 0.8 to 0.93, indicating they are not interchangeable. After reclassification, the proportion of normal scan rates increased in MFR (up to 8%) and niFFR (up to 11.4%). Body mass index (BMI) was the only subgroup factor significantly associated with MC-NMC differences in stress MBF and MFR. CONCLUSIONS: The frame-wise directional motion pattern leads to under- and overestimation. MC influences the systematic bidirectional changes and alters MFR/niFFR classification, making MC and NMC datasets non-interchangeable, which suggests a reduction in false-positive/negative results. </p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13265444/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147766203","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Q.Clear reconstruction shortening the scanning time for ¹¹C-CFT brain PET/MR imaging.","authors":"Leiyue Chen, Wenjie Gong, Dan Wang, He Long, Xin Li, Renxiang Xia, Yun Yao, Wen Zhou, Weicheng Zhou, Xiao Chen","doi":"10.1186/s40658-026-00878-0","DOIUrl":"10.1186/s40658-026-00878-0","url":null,"abstract":"<p><p>PURPOSE: As a tracer in brain PET imaging for diagnosis of Parkinson’s disease (PD), the application of 11C-methyl-N-2β-carbomethoxy-3β-(4-fluorophenyl)-tropane (11C-CFT) is limited due to its short half-life. This study aims to explore the feasibility of using Q.Clear reconstruction to reduce the scanning time of 11C-CFT brain PET imaging. METHODS: We retrospectively enrolled 43 patients who underwent 11C-CFT PET/MRI examination. Their image data was reconstructed using classical ordered subset expectation maximization (OSEM) with 20 min and Q.Clear with 5, 10, 15, 20 min of scanning time (QC05, QC10, QC15, QC20). Subsequently, subjective quality assessment and semi-quantitative assessment were employed across the five reconstruction groups. Subjective image assessment included overall image quality, image noise and lesion depiction. Signal-to-noise ratio (SNR), signal-to-background ratio (SBR) and dopamine transporters (DAT) binding rate were measured in semi-quantitative assessment. RESULTS: In subjective assessment, all Q.Clear groups showed image quality and noise comparable or superior to OSEM. However, QC05 group received significantly lower overall image quality scores than QC10, QC15, QC20 (all P < 0.05). In several QC05 cases, both readers noted conspicuous signal absence with locational and uptake inconsistencies versus OSEM, leading to diagnostic discrepancy. By contrast, QC10, QC15, QC20 provided lesion visibility equivalent to the OSEM reference. Semi‑quantitative analysis indicated no significant differences among all groups in SNR, SBR, or DAT binding rate (all P > 0.05). CONCLUSION: The Q.Clear reconstruction demonstrates reliable performance in the quantitative assessment of 11C-CFT brain PET imaging, therefore can shorten scanning time to at least 10 min under the premise of meeting clinical diagnostic requirements. </p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13243153/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147766173","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-based synthetic CT attenuation correction enables reliable quantitative SPECT in unilateral condylar hyperplasia.","authors":"Anna Rebeka Kovács, Enikő Fanni Juhász, Péter Czina, Ádám Budai, József Varga, Borbála Husztik, Ákos Kovács, Melinda Szoliková, Róbert Boda, Kincső Sára Kovács, Ildikó Garai, Sándor Kristóf Barna","doi":"10.1186/s40658-026-00879-z","DOIUrl":"10.1186/s40658-026-00879-z","url":null,"abstract":"<p><strong>Purpose: </strong>Unilateral condylar hyperplasia (UCH) is a rare mandibular growth disorder in which accurate assessment of condylar metabolic activity is essential for surgical decision-making. Quantitative ^99mTc-methylene diphosphonate (MDP) SPECT/CT is commonly used for this purpose; however, CT-based attenuation correction (CTAC) increases radiation exposure and may be affected by registration errors. Artificial intelligence (AI)-generated synthetic CT (SyCT) has been proposed as a CT-independent alternative. This study aimed to evaluate the agreement between SyCT-based attenuation correction (SyCTAC) and conventional CTAC in quantitative SPECT imaging of UCH.</p><p><strong>Methods: </strong>This retrospective study included 14 patients with UCH who underwent ^99mTc-MDP SPECT/CT. SPECT images were reconstructed with identical parameters using either CTAC or AI-generated SyCTAC. Spherical volumes of interest were placed over both mandibular condyles and the clivus, and maximum and mean standardized uptake values were measured. Relative uptake fractions normalized to the summed condylar activity were calculated for the affected side, while condylar uptake normalized to the clivus was evaluated for both sides. Agreement between CTAC- and SyCTAC-derived indices was assessed using Bland-Altman analysis and linear mixed-effects models.</p><p><strong>Results: </strong>Visual evaluation revealed no relevant differences in image quality between the two reconstructions. Relative uptake fractions normalized to the summed condylar activity showed good agreement between methods. Clivus-normalized ratios demonstrated a small positive bias.</p><p><strong>Conclusion: </strong>AI-generated SyCT provides attenuation correction comparable to CTAC for clinically relevant relative condylar uptake assessment in UCH, supporting its use as a low-radiation alternative for quantitative mandibular SPECT imaging.</p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-04-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13230450/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147728649","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effective elimination of respiratory misregistration-induced attenuation correction errors in PET/CT via deep learning trained on data-driven gated PET from strictly respiratory-phase-matched PET/CT cases.","authors":"Koki Takamura, Koichi Ishizu, Chio Okuyama, Tomohiro Ueno, Naozo Sugimoto","doi":"10.1186/s40658-026-00880-6","DOIUrl":"10.1186/s40658-026-00880-6","url":null,"abstract":"<p><p>BACKGROUND: In positron emission tomography (PET)/computed tomography (CT), CT is used for attenuation correction (AC). CT-based AC (CTAC) is susceptible to misregistration due to respiratory phase differences between CT and PET, frequently resulting in a “banana artifact (BA)”—an underestimation of tracer uptake immediately below the diaphragm. This study aimed to evaluate a deep learning-based CT-less AC (DLAC), trained on respiratory-phase-matched data using data-driven gated (DDG) PET, for its efficacy in reducing BA. METHODS: We retrospectively analyzed 18F-fluorodeoxyglucose (FDG) PET/CT datasets from a Discovery MI-25 (GE Healthcare) with DDG PET acquisition. DDG removed respiratory blurring in PET, and no AC errors were observed in cases with respiratory-phase-matched CT. Out of 1137 consecutive clinical cases, 255 well-aligned cases (WA-cases) and 387 cases with BA (BA-cases) were selected. The WA-cases showed no detectable misregistration in the upper abdomen through visual discrimination. The model, trained on pairs of non-AC PET (NAC-PET) and CTAC PET (CTAC-PET) of the WA-cases, processed NAC-PET to generate DLAC PET (DLAC-PET), resulting in reduced AC errors. Performance was assessed via fivefold cross-validation in WA-cases and artifact reduction testing in BA-cases. RESULTS: In WA-cases, DLAC-PET highly correlated with CTAC-PET, with metrics showing excellent agreement: mean absolute error (MAE) 0.029, root mean squared error (RMSE) 0.086, peak signal-to-noise ratio (PSNR) 35.5 dB, structural similarity (SSIM) 0.937, and Pearson correlation coefficient (r) 0.984. MAE and RMSE are expressed in standardized uptake values units. In BA-cases, DLAC-PET successfully suppressed BA in 385 out of 387 cases (99.5%), and no BA was observed in 366 cases (94.6%). CONCLUSIONS: Our DLAC model, trained on DDG-derived respiratory-phase-matched NAC-PET and CTAC-PET, effectively and consistently reduced BA in almost all problematic cases, potentially facilitating clinical interpretation in CTAC-failing cases. </p>","PeriodicalId":11559,"journal":{"name":"EJNMMI Physics","volume":" ","pages":""},"PeriodicalIF":3.5,"publicationDate":"2026-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13222924/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147722225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}