RATIONALE AND OBJECTIVES:To explore the accuracy, reliability, and diagnostic performance of photon-counting CT (PCCT) spectral localizer radiographs (SLRs) for quantifying lumbar areal bone mineral density (BMD) and detecting osteoporosis (T-score ≤-2.5). MATERIALS AND METHODS:This prospective study recruited consecutive participants from April to July 2024. Participants each underwent a dual-energy X-ray absorptiometry (DXA) examination serving as the gold-standard reference for aBMD (aBMDDXA) and a PCCT scan to obtain SLR. The SLRs were reconstructed into hydroxyapatite (HA) and water maps. Lumbar vertebrae (L1 to L4) and soft tissue were blindly and semiautomatically segmented on HA and water maps to calculate aBMDSLR. The agreement and relative absolute error (RAE) between aBMDSLR and aBMDDXA were calculated. Factors that might influence the RAE were evaluated. Using DXA results as the reference, the diagnostic performance of PCCT-SLRs for osteoporosis was assessed. RESULTS:A total of 159 participants (88 females) with a median age of 66 years (interquartile range [IQR], 55-72 years) were included. The median (IQR) aBMDDXA and aBMDSLR values were 1.095 (0.936-1.261) g/cm2 and 1.086 (0.932-1.255) g/cm2, respectively. There was excellent agreement between the two methods (mean bias=-0.57%). The median (IQR) RAE was 2.65% (1.23-4.07%). The RAE was unaffected by age, body mass index, aBMD, sex, tube voltage, or tube current. The sensitivity and specificity of PCCT-SLRs for osteoporosis diagnosis were 92.31% (12/13) and 98.63% (144/146), respectively. CONCLUSION:The PCCT-SLR is an accurate and reliable approach for lumbar aBMD quantification in humans, with high diagnostic performance for osteoporosis.

To assess the impact of acquisition and reconstruction factors on the robustness of radiomics within photon-counting detector CT (PCD-CT). A phantom with twenty-eight texture materials was scanned with different acquisition and reconstruction factors including reposition, scan mode (standard vs high-pitch), tube voltage (120 kVp vs 140 kVp), slice thickness (1.0 mm vs 0.4 mm), radiation dose level (0.5 mGy, 1.0 mGy, 3.0 mGy, 5.0 mGy, vs 10.0 mGy), quantum iterative reconstruction level (0/4, 2/4, vs 4/4), and reconstruction kernel (Qr40, Qr44, vs Qr48). Thirteen sets of virtual monochromatic images at 70-keV were reconstructed. The regions of interest were drawn with rigid registrations. Ninety-three radiomics features were extracted from each material. The reproducibility of radiomics features was evaluated using the intraclass correlation coefficient (ICC) and concordance correlation coefficient (CCC). The variability of radiomics features was assessed by coefficient of variation (CV) and quartile coefficient of dispersion (QCD). The percentage of features with ICC > 0.90 and CCC > 0.90 were high when repositioned (88.2

To clarify the fat attenuation index (FAI) change trend of peri-saphenous vein graft (SVG) and determine the association between FAI and graft disease progression based on CCTA images. Patients with venous coronary artery bypass grafts (CABGs) were consecutively enrolled in this retrospective study. In study 1, 72 patients who had undergone 1, 3, and 5 years of CCTA examinations without graft occlusion were recruited, and generalized estimation equation was used to analyze the peri-SVG FAI change trend over time. In study 2, 42 patients with graft disease progression and 84 patients as controls were propensity score-matched. Generalized linear mixed model and continuous net reclassification improvement (NRI) were used for assessing the associations with graft disease progression. Multivariable Cox regression analysis was used for assessing risk factors predicting cardiac events. In study 1, both the FAI of proximal right coronary artery and SVG decreased over time. In study 2, the 1-year CTA-derived FAI of grafts and graft anastomosis were independent indicators of graft disease progression at the 3-year CCTA follow-up (graft: odds ratio [OR] = 1.106; 95

BACKGROUND. Head and neck CTA requires fine-detail evaluation, including characterization of potentially very small vessels and intrastent lumina. Blooming artifacts also hinder evaluation. OBJECTIVE. The purpose of this study was to evaluate image quality of ultrahigh-resolution (UHR) photon-counting detector (PCD) CTA of the head and neck and to explore variation of image quality across body vascular (Bv) reconstruction kernels. METHODS. This prospective study included patients who underwent clinically indicated head and neck CTA from September 2023 to December 2023. Participants underwent PCD CTA in the UHR mode. Reconstructions for each examination included a reference reconstruction (reflecting clinical protocols) using a 0.8-mm slice thickness and Bv40 kernel and six UHR reconstructions using a 0.2-mm slice thickness and kernels of varying sharpness (Bv48-Bv80). Quantitative measures were recorded. Two radiologists independently evaluated qualitative measures using a Likert scale ranging from 1 (lowest quality) to 5 (highest quality). RESULTS. The analysis included 103 participants (mean age, 61.3 ± 13.0 [SD] years; 56 men, 47 women). Median vessel sharpness (in Hounsfield units per millimeter) was 100.9 HU/mm for the reference reconstruction, and for the UHR reconstructions, it varied from 110.0 HU/mm for Bv48 to 121.6 HU/mm for Bv76 and 134.7 HU/mm for Bv80. Median right internal carotid artery C2 luminal diameter was 3.8 mm for the reference reconstruction, and for the UHR reconstructions, it increased from 4.1 mm for Bv48 to 4.9 mm for Bv80. For both readers, median overall image quality for the reference reconstruction was 3, and for the UHR reconstructions, it was highest for Bv64 (score of 5); calcified-plaque blooming artifact for the reference reconstruction was 1, and for the UHR reconstructions, it was highest for Bv72 (score of 5) and Bv76 (score of 5); stent blooming artifact for the reference reconstruction was 1, and for the UHR reconstructions, it was highest for Bv76 (score of 5) and Bv80 (score of 5); soft-plaque delineation for the reference reconstruction was 1, and for the UHR reconstructions, it was highest for Bv76 (score of 5) or Bv80 (score of 5); small-vessel visualization for the reference reconstruction was 1, and for the UHR reconstructions, it was highest for Bv76 (score of 5) or Bv80 (score of 5). CONCLUSION. UHR PCD CTA yielded reduced blooming artifact from calcified plaques or stents and improved visualization of soft plaque and small vessels. These advantages were more pronounced for strongest kernels, although subjective image quality was better for a weaker kernel. CLINICAL IMPACT. The findings indicate benefits from the use of UHR PCD CTA for head and neck evaluation and may help guide such examinations' kernel selection.

Rationale and Objectives: To evaluate the feasibility of using photon-counting detector computed tomography (PCD CT) to simultaneously quantify fat and iron content Materials and Methods: Phantoms with pure fat, pure iron and fat-iron deposition were scanned by two tube voltages (120 and 140 kV) and two image quality (IQ) settings (80 and 145). Using an iron-specific three-material decomposition algorithm, virtual noniron (VNI) and virtual iron content (VIC) images were generated at quantum iterative reconstruction (QIR) strength levels 1-4. Results: Significant linear correlations were observed between known fat content (FC) and VNI for pure fat phantoms (r = 0.981-0.999, p < 0.001) and between known iron content (IC) and VIC for pure iron phantoms (r = 0.897-0.975, p < 0.001). In fat-iron phantoms, the measurement for fat content of 5-30% demonstrated good linearity between FC and VNI (r = 0.919-0.990, p < 0.001), and VNI were not affected by 75, 150, and 225 mol/g iron overload (p = 0.174-0.519). The measurement for iron demonstrated a linear range of 75-225 mu mol/g between IC and VIC (r = 0.961-0.994, p < 0.001) and VIC was not confounded by the coexisting 5%, 20%, and 30% fat deposition (p = 0.943-0.999). The Bland-Altman of fat and iron measurements were not significantly different at varying tube voltages and IQ settings (all p > 0.05). No significant difference in VNI and VIC at QIR 1-4. Conclusion: PCD CT can accurately and simultaneously quantify fat and iron, including scan parameters with lower radiation dose.

Background Traditional energy-integrating detector CT has limited utility in accurately quantifying liver fat due to protocol-induced CT value shifts, but this limitation can be addressed by using photon-counting detector (PCD) CT, which allows for a standardized CT value. Purpose To develop and validate a universal CT to MRI fat conversion formula to enhance fat quantification accuracy across various PCD CT protocols relative to MRI proton density fat fraction (PDFF). Materials and Methods In this prospective study, the feasibility of fat quantification was evaluated in phantoms with various nominal fat fractions. Five hundred asymptomatic participants and 157 participants with suspected metabolic dysfunction-associated steatotic liver disease (MASLD) were enrolled between September 2023 and March 2024. Participants were randomly assigned to six groups with different CT protocols regarding tube voltage (90, 120, or 140 kVp) and radiation dose (standard or low). Of the participants in the 120-kVp standard-dose asymptomatic group, 51% (53 of 104) were designated as the training cohort, with the rest of the asymptomatic group serving as the validation cohort. A CT to MRI fat quantification formula was derived from the training cohort to estimate the CT-derived fat fraction (CTFF). CTFF agreement with PDFF and its error were evaluated in the asymptomatic validation cohort and subcohorts stratified by tube voltage, radiation dose, and body mass index, and in the MASLD cohort. The factors influencing CTFF error were further evaluated. Results In the phantoms, CTFF showed excellent agreement with nominal fat fraction (intraclass correlation coefficient, 0.98; mean bias, 0.2%). A total of 412 asymptomatic participants and 122 participants with MASLD were included. A CT to MRI fat conversion formula was derived as follows: MRI PDFF (%) = -0.58 · CT (HU) + 43.1. Across all comparisons, CTFF demonstrated excellent agreement with PDFF (mean bias values < 1%). CTFF error was not influenced by tube voltage, radiation dose, body mass index, or PDFF. Agreement between CTFF and PDFF was also found in the MASLD cohort (mean bias, -0.2%). Conclusion Standardized CT value from PCD CT showed a robust and remarkable agreement with MRI PDFF across various protocols and may serve as a precise alternative for liver fat quantification. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Wildman-Tobriner in this issue.

To develop and validate deep learning (DL)-models that denoise late iodine enhancement (LIE) images and enable accurate extracellular volume (ECV) quantification. This study retrospectively included patients with chest discomfort who underwent CT myocardial perfusion + CT angiography + LIE from two hospitals. Two DL models, residual dense network (RDN) and conditional generative adversarial network (cGAN), were developed and validated. 423 patients were randomly divided into training (182 patients), tuning (48 patients), internal validation (92 patients) and external validation group (101 patients). LIEsingle (single-stack image), LIEaveraging (averaging multiple-stack images), LIERDN (single-stack image denoised by RDN) and LIEGAN (single-stack image denoised by cGAN) were generated. We compared image quality score, signal-to-noise (SNR) and contrast-to-noise (CNR) of four LIE sets. The identifiability of denoised images for positive LIE and increased ECV (> 30

We aimed to evaluate the reproducibility of computed tomography (CT) radiomic features (RFs) about Epicardial Adipose Tissue (EAT). The features derived from coronary photon-counting computed tomography (PCCT) angiography datasets using the PureCalcium (VNCPC) and conventional virtual non-contrast (VNCConv) algorithm were compared with true non-contrast (TNC) series. RFs of EAT from 52 patients who underwent PCCT were quantified using VNCPC, VNCConv, and TNC series. The agreement of EAT volume (EATV) and EAT density (EATD) was evaluated using Pearson’s correlation coefficient and Bland-Altman analysis. A total of 1530 RFs were included. They are divided into 17 feature categories, each containing 90 RFs. The intraclass correlation coefficients (ICCs) and concordance correlation coefficients (CCCs) were calculated to assess the reproducibility of RFs. The cutoff value considered indicative of reproducible features was > 0.75. the VNCPC and VNCConv tended to underestimate EATVs and overestimate EATDs. Both EATV and EATD of VNCPC series showed higher correlation and agreement with TNC than VNCConv series. All types of RFs from VNCPC series showed greater reproducibility than VNCConv series. Across all image filters, the Square filter exhibited the highest level of reproducibility (ICC = 67/90, 74.4

OBJECTIVES:To investigate the predictive value of coronary computed tomography angiography-derived fractional flow reserve (CT-FFR) before percutaneous coronary intervention (PCI) to predict target vessel failure (TVF) after stent implantation. METHODS:This retrospective study included 429 patients (429 vessels) who underwent PCI and stent implantation after CCTA within 3 months. All patients underwent coronary stent implantation between January 2012 and December 2019. A dedicated workstation (Syngo Via, Siemens) was used to analyze and measure the CT-FFR value. The cut-off values of pre-PCI CT-FFR for predicting TVF were defined as 0.80 and the value using the log-rank maximization method, respectively. The primary outcome was TVF, defined as a composite of cardiac death, target vessel myocardial infarction, and clinically driven target vessel revascularization (TVR), which was a secondary outcome. RESULTS:During a median 64.0 months follow-up, the cumulative incidence of TVF was 7.9% (34/429). The cutoff value of pre-PCI CT-FFR based on the log-rank maximization method was 0.74, which was the independent predictor for TVF [hazard ratio (HR): 2.61 (95% CI: 1.13, 6.02); P =0.024] and TVR [HR: 3.63 (95%CI: 1.25, 10.51); P =0.018]. Compared with the clinical risk factor model, pre-PCI CT-FFR significantly improved the reclassification ability for TVF [net reclassification improvement (NRI), 0.424, P <0.001; integrative discrimination index (IDI), 0.011, P =0.022]. Adding stent information to the prediction model resulted in an improvement in reclassification for the TVF (C statistics: 0.711, P =0.001; NRI: 0.494, P <0.001; IDI: 0.020, P =0.028). CONCLUSIONS:Pre-PCI CT-FFR ≤0.74 was an independent predictor for TVF or TVR, and integration of clinical, pre-PCI CT-FFR, and stent information models can provide a better risk stratification model in patients with stent implantation.

The Fitzgibbon classification and Coronary Artery Disease—Reporting and Data System (CAD-RADS) were the main assessment methods for graft patency in post-coronary artery bypass graft (CABG) patients. Yet they only included the stenosis degree and have limited predictive power for long-term outcomes. We thus developed a novel scoring system (graft score) for graft evaluation in post-CABG patients by integrating stenosis degree, vessel weight coefficient, and native artery lesions based on coronary computed tomography angiography (CCTA) for guiding and optimizing post-CABG treatment. This retrospective study enrolled consecutive post-CABG patients indicated for 1-year follow-up CCTA examination. The primary outcome was major adverse cardiovascular event (MACE). Multivariable Cox regression analysis, Harrell’s C-statistics, and continuous net reclassification were used for statistical analyses. A total of 402 patients (median age: 64 years, interquartile range: 58–69; 81.19

To investigate the clinical value of dual-source dual-energy CT (dsDECT) quantitative parameters in evaluating hemodynamics and predicting high-risk gastroesophageal varices in cirrhotic patients. 98 consecutive patients were collected in this prospectively study and all patients underwent an abdominal triple-phase contrasted-enhanced examination with dsDECT. Iodine concentration (IC) and normalized iodine concentration (NIC) of the liver parenchyma, spleen parenchyma and aorta at different phases were recorded, and arterial iodine fraction (AIF), iodine washout rate (IWR), and extracellular volume (ECV) were calculated. Using upper gastrointestinal endoscopy as the reference standard, patients who met the inclusion and exclusion criteria were divided into groups with varices need treatment (VNT) and non-VNT. The clinical characteristics, traditional CT features and quantitative dsDECT parameters were compared between the VNT group and the non-VNT group using univariate analysis. The binary logistics analysis was used to build a model for diagnosing VNT. The receiver operating characteristic (ROC) curve was used for analysis and the DeLong test was used to compare different ROC curves. Finally, 57 patients were included in this study. Univariate analysis showed statistically significant differences in NIC of the liver at the portal venous phase (NIC-LPVP), IWR of the liver (IWR-L) and spleen volume between the VNT group and the non-VNT group (p < 0.05). The mixed-CT model was built by binary logistics analysis. The ROC curves of NIC-LPVP, IWR-L, spleen volume and the mixed-CT model were statistically significant (p < 0.05) for predicting VNT in cirrhotic patients, among which the area under the ROC curve of the mixed-CT model was the highest. Dual-source dual-energy CT has added clinical value in evaluating hepatic hemodynamics and diagnosing VNT in patients with liver cirrhosis.

BackgroundTo construct and validate the CT-based radiomics model for predicting the tyrosine kinase inhibitors (TKIs) effects in osteosarcoma (OS) patients with pulmonary metastasis.MethodsOS patients with pulmonary metastasis treated with TKIs were randomly separated into training and testing cohorts (2:1 ratio). Radiomic features were extracted from the baseline unenhanced chest CT images. The random survival forest (RSF) and Kaplan-Meier survival analyses were performed to construct and evaluate radiomics signatures (R-model-derived). The univariant and multivariant Cox regression analyses were conducted to establish clinical (C-model) and combined models (RC-model). The discrimination abilities, goodness of fit and clinical benefits of the three models were assessed and validated in both training and testing cohorts.ResultsA total of 90 patients, 57 men and 33 women, with a mean age of 18 years and median progression-free survival (PFS) of 7.2 months, were enrolled. The R-model was developed with nine radiomic features and demonstrated significant predictive and prognostic values. In both training and testing cohorts, the time-dependent area under the receiver operating characteristic curves (AUC) of the R-model and RC-model exhibited obvious superiority over C-model. The calibration and decision curve analysis (DCA) curves indicated that the accuracy of the R-model was comparable to RC-model, which exhibited significantly better performance than C-model.ConclusionsThe R-model showed promising potential as a predictor for TKI responses in OS patients with pulmonary metastasis. It can potentially identify pulmonary metastatic OS patients most likely to benefit from TKIs treatment and help guide optimized clinical decisions.

This study aimed to compare the image quality and diagnostic performance of standard-resolution (SR) and ultra-high-resolution (UHR) coronary CT angiography (CCTA) based on photon-counting detector CT (PCD-CT) of coronary stents and explore the best reconstruction kernel for stent imaging. From July 2023 to September 2023, patients were enrolled to undergo CCTA using a dual-source PCD-CT system after coronary angioplasty with stent placement. SR images with a slice thickness/increment of 0.6/0.4 mm were reconstructed using a vascular kernel (Bv48), while UHR images with a slice thickness/increment of 0.2/0.2 mm were reconstructed using vascular kernels of six sharpness levels (Bv48, Bv56, Bv60, Bv64, Bv72, and Bv76). The in-stent lumen diameters were evaluated. Subjective image quality was also evaluated by a 5-point Likert scale. Invasive coronary angiography was conducted in 12 patients (25 stents). Sixty-nine patients (68.0 [61.0, 73.0] years, 46 males) with 131 stents were included. All UHR images had significantly larger in-stent lumen diameter than SR images (p < 0.001). Specifically, UHR-Bv72 and UHR-Bv76 for in-stent lumen diameter (2.17 [1.93, 2.63] mm versus 2.20 [1.93, 2.59] mm) ranked the two best kernels. The subjective analysis demonstrated that UHR-Bv72 images had the most pronounced effect on reducing blooming artifacts, showcasing in-stent lumen and stent demonstration, and diagnostic confidence (p < 0.001). Furthermore, SR and UHR-Bv72 images showed a diagnostic accuracy of 78.3

Objective: Fat deposition is an important marker of many metabolic diseases. As a noninvasive and convenient examination method, CT has been widely used for fat quantification. With the clinical application of photon-counting detector (PCD)-CT, we aimed to investigate the accuracy, stability, and dose level of PCD-CT using various scan settings for fat quantification. Materials and Methods: Eleven agar-based lipid-containing phantoms (vials with different fat fractions [FFs]; range: 0 % -100 %) were scanned using PCD-CT. Three scanning types (sequence scan, regular spiral scan with a pitch of 0.8, and high-pitch spiral scan with a pitch of 3.2), four tube voltages (90, 120, 140, and 100 kV with a tin filter), and three image quality (IQ) levels (IQ levels of 20, 40, and 80) were alternated, and each scan setting was used twice. For each scan, a 70-keV image was generated using the same reconstruction parameters. A regular spiral scan at 120 kV with IQ80 was used to transfer the CT numbers of all scans to the FF. Intraclass correlation coefficient (ICC) and Bland -Altman analysis were implemented for accuracy and agreement evaluation, and group differences were compared using analysis of variance. Results: Excellent agreement and accuracy of FF derived by PCD-CT with all scan settings was demonstrated by high ICCs ( >0.9; range: 0.929 -0.998, p < 0.017) and low bias ( <5% range: -2.9 % -5%). The root mean square error (RMSE) between the PCD-CT-acquired FF and the reference standard ranged from 1.0 % to 5.0 %, among which the high-pitch scan at 120 kV with IQ20 accounted for the lowest RMSE (1.0 %). The spiral scan at 120 kV with IQ20 and IQ80 yielded the lowest bias (mean value: 1.19 % and 1.23 %, respectively). Conclusion: Fat quantification using PCD-CT reconstructed at 70 keV was accurate and stable under various scan settings. PCD-CT has great potential for fat quantification using ultralow radiation doses.

This study aimed to investigate the potential benefit of ultra-high-resolution (UHR) photon-counting detector CT (PCD-CT) angiography in visualization of abdominal arteries in comparison to standard-reconstruction (SR) images of virtual monoenergetic images (VMI) at low kiloelectron volt (keV). We prospectively included 47 and 47 participants to undergo contrast-enhanced abdominal CT scans within UHR mode on a PCD-CT system using full-dose (FD) and low-dose (LD) protocols, respectively. The data were reconstructed into six series of images: FD_UHR_Br48, FD_UHR_Bv56, FD_UHR_Bv60, FD_SR_Bv40, LD_UHR_Bv48, and LD_SR_Bv40. The UHR reconstructions were performed with three kernels (Bv48, Bv56, and Bv60) within 0.2 mm. The SR were virtual monoenergetic imaging reconstruction with Bv40 kernel at 40-keV within 1 mm. Each series of axial images were reconstructed into coronal and volume-rendered images. The signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of seven arteries were measured. Three radiologists assessed the image quality, and visibility of nine arteries on all the images. SNR and CNR values of SR images were significantly higher than those of UHR images (P < 0.001). The SR images have higher ratings in image noise (P < 0.001), but the FD_UHR_Bv56 and FD_UHR_Bv60 images has higher rating in vessel sharpness (P < 0.001). The overall quality was not significantly different among FD_VMI_40keV, LD_VMI_40keV, FD_UHR_Bv48, and LD_UHR_Bv48 images (P > 0.05) but higher than those of FD_UHR_Bv56 and FD_UHR_Bv60 images (P < 0.001). There is no significant difference of nine abdominal arteries among six series of images of axial, coronal and volume-rendered images (P > 0.05). To conclude, 1-mm SR image of VMI at 40-keV is superior to 0.2-mm UHR regardless of which kernel is used to visualize abdominal arteries, while 0.2-mm UHR image using a relatively smooth kernel may allow similar image quality and artery visibility when thinner slice image is warranted.

Aims: To investigate the long-term prognostic value of coronary computed tomography angiography (CCTA)-derived high-risk attributes and radiomic features of pericoronary adipose tissue (PCAT) in diabetic patients for predicting major adverse cardiac event (MACE). Methods and Results: Diabetic patients with intermediate pre-test probability of coronary artery disease were prospectively enrolled and referred for CCTA. Three models (model-1 with clinical parameters; model-2 with clinical factors + CCTA imaging parameters; model-3 with the above parameters and PCAT radiomic features) were developed in the training cohort (835 patients) and tested in the independent validation cohort (557 patients). 1392 patients were included and MACEs occurred in 108 patients (7.8%). Multivariable Cox regression analysis revealed that HbA1c, coronary calcium Agatston score, significant stenosis and high-risk plaque were independent predictors for MACE whereas none of PCAT radiomic features showed predictive value. In the training cohort, model-2 demonstrated higher predictive performance over model-1 (C-index = 0.79 vs. 0.68, p < 0.001) whereas model-3 did not show incremental value over model-2(C-index = 0.79 vs. 0.80, p = 0.408). Similar findings were found in the validation cohort. Conclusions: The combined model (clinical and CCTA high-risk anatomical features) demonstrated high efficacy in predicting MACE in diabetes. PCAT radiomic features failed to show incremental value for risk stratification. (c) 2024 The Royal College of Radiologists. Published by Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Abstract Objectives To develop and validate nomograms combining radiomics and semantic features to identify the invasiveness and histopathological risk stratification of thymic epithelial tumors (TET) using contrast-enhanced CT. Methods This retrospective multi-center study included 224 consecutive cases. For each case, 6764 intratumor and peritumor radiomics features and 31 semantic features were collected. Multi-feature selections and decision tree models were performed on radiomics features and semantic features separately to select the most important features for Masaoka–Koga staging and WHO classification. The selected features were then combined to create nomograms for the two systems. The performance of the radiomics model, semantic model, and combined model was evaluated using the area under the receiver operating characteristic curves (AUCs). Results One hundred eighty-seven cases (56.5 years ± 12.3, 101 men) were included, with 62 cases as the external test set. For Masaoka–Koga staging, the combined model, which incorporated five peritumor radiomics features and four semantic features, showed an AUC of 0.958 (95% CI: 0.912–1.000) in distinguishing between early-stage (stage I/II) and advanced-stage (III/IV) TET in the external test set. For WHO classification, the combined model incorporating five peritumor radiomics features and two semantic features showed an AUC of 0.857 (0.760–0.955) in differentiating low-risk (type A/AB/B1) and high-risk (B2/B3/C) TET. The combined models showed the most effective predictive performance, while the semantic models exhibited comparable performance to the radiomics models in both systems (p > 0.05). Conclusion The nomograms combining peritumor radiomics features and semantic features could help in increasing the accuracy of grading invasiveness and risk stratification of TET. Critical relevance statement Peripheral invasion and histopathological type are major determinants of treatment and prognosis of TET. The integration of peritumoral radiomics features and semantic features into nomograms may enhance the accuracy of grading invasiveness and risk stratification of TET. Key Points Peritumor region of TET may suggest histopathological and invasive risk. Peritumor radiomic and semantic features allow classification by Masaoka–Koga staging (AUC: 0.958). Peritumor radiomic and semantic features enable the classification of histopathological risk (AUC: 0.857). Graphical Abstract

To investigate the feasibility and accuracy of iodine quantification using PCD-CT in standard-pitch and high-pitch scanning at different scan parameters in a phantom model. Four inserts with known iodine concentrations (2, 5, 10, and 15 mg/mL) were placed in the removable CT phantom and scanned using high-pitch (3.2) and standard-pitch (0.8) modes on PCD-CT. Two tube voltages (120 and 140 kVp) and four radiation doses (1, 3, 5, and 10 mGy) were alternated. Each scan setting was repeated three times. Mean iodine concentration for each insert across three consecutive slices was recorded. Percentage absolute bias (PAB) was assessed for iodine quantification. A total of 96 acquisitions were conducted. In small phantom, the average for PAB was 2.96

To validate the accuracy of coronary artery calcium score (CACS) using photon-counting detector (PCD) CT under various scanning settings and explore the optimized scanning settings considering both the accuracy and the radiation dose. A CACS phantom containing six hollow cylindrical hydroxyapatite calcifications of two sizes with three densities and 12 patients underwent CACS scans. For PCD-CT, two scanning modes (sequence and flash [high-pitch spiral mode]) and five tube voltages (90kV, 120kV, 140kV, Sn100kV, and Sn140kV) at different image quality (IQ) levels were set for phantom, and patients were scanned with 120kV at IQ19 using flash mode. All acquisitions from PCD-CT were reconstructed at 70keV. Acquisitions in sequence mode at 120kV on an energy-integrating detector CT (EID-CT) was used as the reference. Agatston, mass, and volume scores were calculated. The CACS from PCD-CT exhibited excellent agreements with the reference (all intraclass correlation coefficient [ICC] > 0.99). The root mean square error (RMSE) between the Agatston score acquired from PCD-CT and the reference (5.4–11.5) was small. A radiation dose reduction (16–75

BACKGROUND. Use of virtual monoenergetic images (VMIs) from multienergy CT scans can mitigate inconsistencies in traditional attenuation measurements that result from variation in scan-related factors. Photon-counting detector (PCD) CT systems produce VMIs as standard image output under flexible scanning conditions. OBJECTIVE. The purpose of this article was to evaluate the consistency of monoenergetic attenuation measurements obtained from a clinical PCD CT scanner across a spectrum of scanning paradigms. METHODS. A phantom with 10 tissue-simulating inserts was imaged using a clinical dual-source PCD CT scanner. Nine scanning paradigms were obtained across combinations of tube voltages (90, 120, and 140 kVp) and image quality (IQ) levels (80, 145, and 180). Images were reconstructed at VMI levels of 50, 60, 70, and 80 keV. Consistency of attenuation measurements was assessed, using the 120 kVp with IQ level of 145 scanning paradigm as the reference scan. RESULTS. For all scanning paradigms, attenuation measurements showed intra-class correlation of 0.999 and higher with respect to the reference scan. Across inserts, mean bias relative to the reference scan ranged from -14.9 to 13.6 HU, -2.7 to 1.7 HU, and -3.9 to 3.8 HU at tube voltages of 90, 120, and 140 kVp, respectively; and from -14.9 to 13.6 HU, -6.4 to 3.8 HU, -3.7 to 1.4 HU, and -7.2 to 4.3 HU at VMI levels of 50, 60, 70, and 80 keV, respectively. Thus, mean bias did not exceed 5 HU for any insert at tube potentials of 120 kVp and 140 kVp, nor for any insert at a VMI level of 70 keV. At a VMI level of 50 keV and tube potential of 90 kVp, mean bias exceeded 5 HU for 14 of 30 possible combinations of inserts and scanning paradigms and exceeded 10 HU for four of 30 such combinations. At VMI levels of both 60 and 80 keV, mean bias exceeded 5 HU for only two combinations of inserts and scanning paradigms, all at a tube potential of 90 kVp. CONCLUSION. PCD CT generally provided consistent attenuation measurements across combinations of scanning paradigms and VMI levels. CLINICAL IMPACT. PCD CT may facilitate quantitative applications of CT data in clinical practice.