European Journal of Radiology最新文献

{"title":"Metal artifact reduction from surgical clips for intracranial aneurysms in photon-counting detector CT angiography","authors":"Masahiro Nakashima ,&nbsp;Tatsuya Kawai ,&nbsp;Kazuhisa Matsumoto ,&nbsp;Takatsune Kawaguchi ,&nbsp;Nobuo Kitera ,&nbsp;Seita Watanabe ,&nbsp;Akio Hiwatashi","doi":"10.1016/j.ejrad.2025.112411","DOIUrl":"10.1016/j.ejrad.2025.112411","url":null,"abstract":"<div><h3>Objectives</h3><div>To identify conditions that reduce metal artifacts from surgical clips for intracranial aneurysms while preserving parent vessel visualization in CT angiography (CTA) by using a reconstruction method combining virtual monoenergetic image (VMI) and a metal artifact reduction (MAR) algorithm.</div></div><div><h3>Materials &amp; Methods</h3><div>Patients who underwent CTA after clipping surgery using photon-counting detector CT (PCD-CT) in 2023–2025 were included. Images were reconstructed as conventional polychromatic 120kVp images (T3D) and 40, 70, 100, 130, 160, and 190 keV using VMI; we created images applying the MAR algorithm to each. Hyperdense/ hypodense artifacts, the artifact index (AI), and the contrast-to-noise ratio (CNR) were quantitatively measured by two neuroradiologists. Hyperdense/ hypodense artifacts, noise, parent vessel visualization, blooming, and new artifacts generated with the MAR algorithm were qualitatively assessed by four neuroradiologists.</div></div><div><h3>Results</h3><div>28 patients (median 64 [53–72] years, 15 men) were included. Quantitatively, hyperdense artifacts decreased at ≥ 70 keV (68.4H.U.) versus T3D (82.7) (P &lt; 0.001). Hypodense artifacts and AI decreased at ≥ 100 keV (-2.3H.U. and 21.6) (P &lt; 0.001). CNRs decreased at ≥ 70 keV (20.1) and further decreased when the MAR was applied (13.6) (P &lt; 0.001). Qualitatively, hyperdense/ hypodense artifacts decreased at ≥ 100 keV and further decreased slightly with the MAR (P &lt; 0.001). Noise decreased at ≥ 100 keV. The parent vessel visualization after clipping improved at 70 and 100 keV but decreased with the MAR (P &lt; 0.001). Clip blooming did not occur at ≥ 100 keV (P &lt; 0.001) regardless of the use/nonuse of the MAR (P &gt; 0.500). New artifacts decreased at ≥ 100 keV-MAR versus T3D-MAR (P &lt; 0.001).</div></div><div><h3>Conclusion</h3><div>The optimal setting for reducing metal artifacts while preserving vascular information was 100 keV.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112411"},"PeriodicalIF":3.3,"publicationDate":"2025-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145043984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel multimodal framework combining habitat radiomics, deep learning, and conventional radiomics for predicting MGMT gene promoter methylation in Glioma: Superior performance of integrated models","authors":"Feng-Ying Zhu ,&nbsp;Wen-Jing Chen ,&nbsp;Hao-Yan Chen ,&nbsp;Si-Yu Ren ,&nbsp;Li-Yong Zhuo ,&nbsp;Tian-Da Wang ,&nbsp;Cong-Cong Ren ,&nbsp;Xiao-Ping Yin ,&nbsp;Jia-Ning Wang","doi":"10.1016/j.ejrad.2025.112406","DOIUrl":"10.1016/j.ejrad.2025.112406","url":null,"abstract":"<div><h3>Purpose</h3><div>The present study aimed to develop a noninvasive predictive framework that integrates clinical data, conventional radiomics, habitat imaging, and deep learning for the preoperative stratification of <em>MGMT</em> gene promoter methylation in glioma.</div></div><div><h3>Materials and Methods</h3><div>This retrospective study included 410 patients from the University of California, San Francisco, USA, and 102 patients from our hospital. Seven models were constructed using preoperative contrast-enhanced T1-weighted MRI with gadobenate dimeglumine as the contrast agent. Habitat radiomics features were extracted from tumor subregions by k-means clustering, while deep learning features were acquired using a 3D convolutional neural network. Model performance was evaluated based on area under the curve (AUC) value, F1-score, and decision curve analysis.</div></div><div><h3>Results</h3><div>The combined model integrating clinical data, conventional radiomics, habitat imaging features, and deep learning achieved the highest performance (training AUC = 0.979 [95 % CI: 0.969–0.990], F1-score = 0.944; testing AUC = 0.777 [0.651–0.904], F1-score = 0.711). Among the single-modality models, habitat radiomics outperformed the other models (training AUC = 0.960 [0.954–0.983]; testing AUC = 0.724 [0.573–0.875]).</div></div><div><h3>Conclusion</h3><div>The proposed multimodal framework considerably enhances preoperative prediction of <em>MGMT</em> gene promoter methylation, with habitat radiomics highlighting the critical role of tumor heterogeneity. This approach provides a scalable tool for personalized management of glioma.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112406"},"PeriodicalIF":3.3,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145026641","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reference standard methodology in the clinical evaluation of AI chest X-ray algorithms for lung cancer detection: A systematic review","authors":"Sean F. Duncan ,&nbsp;Andrew C. Kidd ,&nbsp;Jesus Perdomo Lampignano ,&nbsp;Paul Cannon ,&nbsp;Mark Hall ,&nbsp;David B. Stobo ,&nbsp;John D. Maclay ,&nbsp;Kevin G. Blyth ,&nbsp;David J. Lowe","doi":"10.1016/j.ejrad.2025.112409","DOIUrl":"10.1016/j.ejrad.2025.112409","url":null,"abstract":"<div><h3>Background</h3><div>Lung cancer remains the leading cause of cancer death worldwide, with early diagnosis linked to improved survival. Artificial intelligence (AI) holds promise for augmenting radiologists’ workflows in chest X-ray (CXR) interpretation, particularly for detecting thoracic malignancies. However, clinical implementation of this technology relies on robust and standardised reference standard methodology at the patient-level.</div></div><div><h3>Purpose</h3><div>This systematic review aims to describe reference standard methodology in the clinical evaluation of CXR algorithms for lung cancer detection.</div></div><div><h3>Materials and Methods</h3><div>Searches targeted studies on AI CXR analysis across MEDLINE, Embase, CENTRAL, and trial registries. 2 reviewers independently screened titles and abstracts, with disagreements resolved by a 3rd reviewer. Studies lacking external validation in real-world cohorts were excluded. Bias was assessed using a modified QUADAS-2 tool, and data synthesis followed SWiM guidelines.</div></div><div><h3>Results</h3><div>1,679 papers were screened with 46 papers included for full paper review. 24 different AI solutions were evaluated across a broad range of research questions. We identified significant heterogeneity in reference standard methodology, including variations in target abnormalities, reference standard modality, expert panel composition, and arbitration techniques. 25 % of reference standard parameters were inadequately reported. 66 % of included studies demonstrated high risk of bias in at least one domain.</div></div><div><h3>Discussion</h3><div>To our knowledge, this is the first systematic description of patient-level reference standard methodology in CXR AI analysis of thoracic malignancy. To facilitate translational progress in this field, researchers undertaking evaluations of diagnostic algorithms at the patient-level should ensure that reference standards are aligned with clinical workflows and adhere to reporting guidelines. Limitations include a lack of prospective studies.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112409"},"PeriodicalIF":3.3,"publicationDate":"2025-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145079451","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and validation of a radiomics model based on the ASPECTS framework using CT imaging for predicting malignant cerebral edema","authors":"LiJun Huang ,&nbsp;XiaoQuan Xu ,&nbsp;Bing Tian ,&nbsp;AnYu Liao ,&nbsp;LiYing Wang ,&nbsp;Xi Shen ,&nbsp;ZeHong Cao ,&nbsp;XiaoYu Liu ,&nbsp;Shanshan Lu ,&nbsp;JiaNan Li ,&nbsp;Feng Shi ,&nbsp;ChangSheng Zhou ,&nbsp;LongJiang Zhang ,&nbsp;FeiYun Wu ,&nbsp;WuSheng Zhu ,&nbsp;Xing Wei ,&nbsp;XiaoQing Cheng ,&nbsp;GuangMing Lu","doi":"10.1016/j.ejrad.2025.112410","DOIUrl":"10.1016/j.ejrad.2025.112410","url":null,"abstract":"<div><h3>Purpose</h3><div>Accurate delineation of the infarct region on acute-phase Computed Tomography (CT) remains challenging, and radiomics applications in stroke are limited. We aimed to develop and validate a multimodal prediction model for malignant cerebral edema (MCE) using clinical and radiomic features extracted based on the Alberta Stroke Program Early Computed Tomography Score (ASPECTS) framework, eliminating the need for manual infarct segmentation.</div></div><div><h3>Method</h3><div>This multicenter retrospective study included patients with acute ischemic stroke (AIS) from five stroke centers who underwent Non-Contrast Computed Tomography (NCCT) and computed tomography angiography (CTA). Radiomic features were extracted from ASPECTS regions. Clinical, imaging-alone, and fused models were developed using machine learning. Model performance was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC) analysis, accuracy, calibration curves, and decision curve analysis (DCA). Shapley Additive exPlanations (SHAP) was used to interpret feature contributions.</div></div><div><h3>Results</h3><div>A total of 708 patients were included (median age: 67 years; interquartile range (IQR): 60–76; 448 men, 63.3 %). In the training cohort, the fused model (AUC = 0.91, 95 %<em>CI</em>: 0.88–0.95) outperformed the clinical (AUC = 0.72, 95 %<em>CI</em>: 0.66–0.78, <em>p</em> &lt; 0.001) and CTA_Features (AUC = 0.83, 95 %<em>CI</em>: 0.79–0.88, <em>p</em> &lt; 0.001) models. In the internal validation cohort, the fused model (AUC = 0.78, 95 %<em>CI</em>: 0.70–0.87) outperformed the clinical model (AUC = 0.68, 95 %<em>CI</em>: 0.59–0.77, <em>p</em> = 0.030). In the external validation, the fused model (AUC = 0.88, 95 %<em>CI</em>: 0.82–0.94) outperformed the clinical (AUC = 0.71, 95 %<em>CI</em>: 0.59–0.83, <em>p</em> = 0.010) and CT_Features (AUC = 0.70, 95 %<em>CI</em>: 0.56–0.85, <em>p</em> = 0.012) models. SHAP analysis identified ASPECTS, National Institutes of Health Stroke Scale (NIHSS) score, and collateral score (CS) as top predictors. The fused model demonstrated the highest specificity (82.5 %) and accuracy (78.3 %).</div></div><div><h3>Conclusions</h3><div>The fused model integrating clinical, radiological, and radiomic features extracted using the ASPECTS framework, demonstrated superior and generalizable predictive performance for early MCE prediction. As it uses routinely acquired baseline NCCT and CTA with readily available admission variables and avoids manual segmentation through an ASPECTS-based framework, it can be integrated into clinical workflows to enable rapid and consistent MCE risk estimation.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112410"},"PeriodicalIF":3.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145074821","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient T staging in nasopharyngeal carcinoma via deep Learning-Based Multi-Modal classification","authors":"Dili Song ,&nbsp;Xu Han ,&nbsp;Yong Li ,&nbsp;Hong Ye ,&nbsp;Yongguang Cai ,&nbsp;Liqun Chen ,&nbsp;Lujian Xu ,&nbsp;Ying Zou ,&nbsp;Haibo Zhang ,&nbsp;Diping Song","doi":"10.1016/j.ejrad.2025.112407","DOIUrl":"10.1016/j.ejrad.2025.112407","url":null,"abstract":"<div><h3>Background</h3><div>Accurate T staging of nasopharyngeal carcinoma (NPC) is crucial for precision therapeutic strategies. The T staging process is associated with significant challenges, including time consumption and variability among observers. This study aimed to develop an efficient, automated T staging system that supports personalized treatment and optimizes clinical workflows.</div></div><div><h3>Methods</h3><div>A total of 609 NPC patients were included, with 487 in the training cohort and 122 in the validation cohort. We employed a multi-modal learning framework that integrates MRI images and reports. Automatically delineated regions of interest (ROIs) served as masks. A hierarchical classification strategy (DeepTree) addressed complex staging challenges. We utilized Vision Transformer (ViT) to extract visual features and BERT to encode text features. To improve data fusion, we applied a Q-Former to integrate visual and textual information. The performances of the methods were evaluated using accuracy (ACC), the area under the receiver operating characteristic curve (AUC), precision, sensitivity (SEN), and specificity (SPE).</div></div><div><h3>Results</h3><div>Integrating images and text via Q-Former demonstrated superior performance overall, significantly surpassing single-modality methods. IT-DTM-BLIP2 demonstrated strong performance, achieving an accuracy (ACC) of 0.787 (95% CI 0.714–0.860). The area under the receiver operating characteristic curve (AUC) values were AUC1 (T2 vs. T3/T4) at 0.815 (0.71–0.900) and AUC2 (T3 vs. T4) at 0.876 (0.782–0.945).</div></div><div><h3>Conclusion</h3><div>Our multi-modal approach consistently performs well, offering a robust automated solution that eliminates the need for manual tumor delineation. This streamlines workflows, reduces subjectivity, and offers decision-making support that may improve workflow efficiency and encourage consistency.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112407"},"PeriodicalIF":3.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145043935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatial heterogeneity and distribution of CT-Based pulmonary vascular volumes in chronic thromboembolic pulmonary hypertension","authors":"Andrew J. Synn ,&nbsp;Pietro Nardelli ,&nbsp;Rahul Renapurkar ,&nbsp;Luisa Quesada ,&nbsp;Gonzalo Vegas Sanchez-Ferrero ,&nbsp;James C. Ross ,&nbsp;Rubén San José Estépar ,&nbsp;Andetta R. Hunsaker ,&nbsp;Aaron B. Waxman ,&nbsp;Jane A. Leopold ,&nbsp;George R. Washko ,&nbsp;Raúl San José Estépar ,&nbsp;Gustavo A. Heresi ,&nbsp;Farbod N. Rahaghi","doi":"10.1016/j.ejrad.2025.112405","DOIUrl":"10.1016/j.ejrad.2025.112405","url":null,"abstract":"<div><h3>Rationale/Objectives</h3><div>Image-based vascular biomarkers may help expedite evaluation of chronic thromboembolic pulmonary hypertension (CTEPH), which remains difficult to diagnose despite available effective therapies. We sought to determine if vascular heterogeneity and central redistribution on chest CT differed between CTEPH, pulmonary arterial hypertension (PAH), and control groups.</div></div><div><h3>Materials/Methods</h3><div>We retrospectively included 108 patients who underwent right heart catheterization and chest CT (2011–2018). Automated CT image analysis was used to calculate volumes of all arteries, all veins, and small arteries/veins (area &lt; 5 mm²). Vascular heterogeneity was assessed by partitioning each lung into isovolumetric segments and calculating coefficients of variation (CV) across segments. Central redistribution was assessed by measuring vascular volumes in central/peripheral lung zones (innermost/outermost fifth, respectively) and calculating central-to-peripheral volume ratios. We constructed multivariable linear regression models to compare vascular heterogeneity and redistribution between CTEPH and control/PAH groups.</div></div><div><h3>Results</h3><div>Of 108 patients, 21 had CTEPH, 47 had PAH, and 40 were controls. For vascular heterogeneity, we found consistently higher CVs (i.e. greater heterogeneity) in CTEPH vs. controls. For small arterial volume, CV was 0.09 units higher (95 % CI: 0.04–0.14, p = 0.0004) in the CTEPH group in adjusted models. Similarly, CVs were higher in CTEPH vs. PAH (p = 0.001). For vascular redistribution, we found greater central redistribution in CTEPH compared to controls/PAH; for small arterial volume, central-to-peripheral ratio was 1.52 units higher in CTEPH vs. controls (95 % CI: 0.78–2.26, p = 0.0001).</div></div><div><h3>Conclusion</h3><div>Volumetric measures of heterogeneity and central distribution of pulmonary vessels can be quantified using CT techniques and may contribute to an image-based signature of CTEPH.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112405"},"PeriodicalIF":3.3,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"End-to-end deep learning model with multi-channel and attention mechanisms for multi-class diagnosis in CT-T staging of advanced gastric cancer","authors":"Bowen Liu ,&nbsp;Pengcheng Jiang ,&nbsp;Zehui Wang ,&nbsp;Xiaoxiao Wang ,&nbsp;Zhixuan Wang ,&nbsp;Chen Peng ,&nbsp;Zhanpeng Liu ,&nbsp;Chao Lu ,&nbsp;Donggang Pan ,&nbsp;Xiuhong Shan","doi":"10.1016/j.ejrad.2025.112408","DOIUrl":"10.1016/j.ejrad.2025.112408","url":null,"abstract":"<div><h3>Background</h3><div>Homogeneous AI assessment is required for CT-T staging of gastric cancer.</div></div><div><h3>Purpose</h3><div>To construct an End-to-End CT-based Deep Learning (DL) model for tumor T-staging in advanced gastric cancer.</div></div><div><h3>Materials and Methods</h3><div>A retrospective study was conducted on 460 cases of presurgical CT patients with advanced gastric cancer between 2011 and 2024. A Three-dimensional (3D)-Convolution (Conv)-UNet based automatic segmentation model was employed to segment tumors, and a SmallFocusNet-based ternary classification model was built for CT-T staging. Finally, these models were integrated to create an end-to-end DL model. The segmentation model’s performance was assessed using the Dice similarity coefficient (DSC), Intersection over Union (IoU) and 95 % Hausdorff Distance (HD_95), while the classification model’s performance was measured with the<!--> <!-->area under the Receiver Operating Characteristic curve (AUC), sensitivity, specificity, and F1-score.Eventually, the end-to-end DL model was compared with the radiologist using the McNemar test.</div></div><div><h3>Results</h3><div>The data were divided into <strong><em>Dataset 1</em></strong>(423 cases for training and test set, mean age, 65.0 years ± 9.46 [SD]) and <strong><em>Dataset 2</em></strong>(37 cases for independent validation set, mean age, 68.8 years ± 9.28 [SD]). For segmentation task, the model achieved a DSC of 0.860 ± 0.065, an IoU of 0.760 ± 0.096 in test set of <strong><em>Dataset 1</em></strong>, and a DSC of 0.870 ± 0.164, an IoU of 0.793 ± 0.168 in <strong><em>Dataset 2</em></strong>. For classification task,the model demonstrated a macro-average AUC of 0.882(95 % <em>CI</em> 0.812–0.926), an average sensitivity of 76.9 % (95 % <em>CI</em> 67.6 %–85.3 %) in test set of <strong><em>Dataset 1</em></strong> and a macro-average AUC of 0.862(95 % <em>CI</em> 0.723–0.942), an average sensitivity of 76.3 % (95 % <em>CI</em> 59.8 %–90.0 %) in <strong><em>Dataset 2</em></strong>. Meanwhile, the DL model’s performance was better than that of radiologist (Accuracy was 91.9 %vs82.1 %, <em>P</em> = 0.007).</div></div><div><h3>Conclusion</h3><div>The end-to-end DL model for CT-T staging is highly accurate and consistent in pre-treatment staging of advanced gastric cancer.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112408"},"PeriodicalIF":3.3,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004261","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Accelerating non-contrast MR angiography of the thoracic aorta using compressed SENSE with deep learning reconstruction","authors":"Jan P. Janssen ,&nbsp;Roman J. Gertz ,&nbsp;Juliana Tristram ,&nbsp;Marvin A. Spurek ,&nbsp;Kenan Kaya ,&nbsp;Robert Terzis ,&nbsp;Robert Hahnfeldt ,&nbsp;Thorsten Gietzen ,&nbsp;David Maintz ,&nbsp;Thorsten Persigehl ,&nbsp;Kilian Weiss ,&nbsp;Lenhard Pennig ,&nbsp;Carsten Gietzen","doi":"10.1016/j.ejrad.2025.112403","DOIUrl":"10.1016/j.ejrad.2025.112403","url":null,"abstract":"<div><h3>Purpose</h3><div>REACT (Relaxation-Enhanced Angiography without ContrasT) is a reliable non-contrast magnetic resonance angiography for imaging of the thoracic aorta but remains time-consuming. This study evaluates acceleration of image acquisition using compressed sensing and parallel imaging (Compressed SENSE, CS) combined with deep learning-based image reconstruction (CS-AI).</div></div><div><h3>Methods</h3><div>In this prospective single-center study, 40 volunteers underwent ECG- and navigator-triggered 3D REACT at 3 T using CS acceleration factor 4 (CS4; reference standard) and 8 (CS8). CS8 data were reconstructed with standard and CS-AI methods (CS8-AI). Two radiologists measured aortic diameters, rated subjective image quality and performed pairwise comparisons. Additionally, objective image quality metrics were calculated.</div></div><div><h3>Results</h3><div>Median scan time was reduced by 44 % (CS4: 8:41 min; CS8/CS8-AI: 4:52 min). All techniques showed excellent agreement in aortic diameter measurements (mean differences &lt; 0.2 mm; <em>P</em> &gt; 0.999). CS8-AI demonstrated reduced mean absolute deviation from CS4 compared to CS8 (0.67 vs. 0.77 mm; <em>P</em> = 0.003), and measurement variance was 40–50 % lower with CS8-AI than with CS8 (inter-/intrarater: <em>P</em> &lt; 0.001), and comparable to CS4. CS8 showed significantly lower subjective image quality scores than CS4 (3.70[3.33–4.00] vs. 4.25[3.90–4.50]; <em>P</em> &lt; 0.001), while CS8-AI showed comparable or higher scores (4.40[4.00–4.70]; P = 0.076). Forced-choice comparisons favored CS4 over CS8 (90 % vs. 2.5 %; <em>P</em> &lt; 0.001), but no preference was observed between CS4 and CS8-AI (42.5 % vs. 37.5 %; <em>P</em> &gt; 0.999). Objective metrics predominantly confirmed the subjective results.</div></div><div><h3>Conclusion</h3><div>Deep learning-based reconstruction enables the acquisition of REACT of the thoracic aorta in less than five minutes while preserving high image quality and maintaining excellent measurement reproducibility.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112403"},"PeriodicalIF":3.3,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145085582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Short occipital circulation time derived from quantitative digital subtraction angiography is associated with headache risk in patients with unruptured brain arteriovenous malformations","authors":"Yong-Sin Hu ,&nbsp;Jr-Wei Wu ,&nbsp;Huai-Che Yang ,&nbsp;Hsiu-Mei Wu ,&nbsp;Cheng-Chia Lee ,&nbsp;Feng-Chi Chang ,&nbsp;Kang-Du Liu ,&nbsp;Chung-Jung Lin","doi":"10.1016/j.ejrad.2025.112402","DOIUrl":"10.1016/j.ejrad.2025.112402","url":null,"abstract":"<div><h3>Purpose</h3><div>To explored key angiographic markers associated with headache risk in patients with unruptured brain arteriovenous malformations (BAVMs).</div></div><div><h3>Methods</h3><div>This retrospective study included patients with unruptured, supratentorial BAVMs without prior interventions who underwent digital subtraction angiography between January 2011 and January 2024. The patients were stratified into headache and nonheadache groups on the basis of symptoms at initial presentation. Patient data and angiograms were analyzed to explore BAVM characteristics and quantitative digital subtraction angiography (QDSA) parameters. Data on occipital circulation time—defined as the interval between the bolus arrival time in the cavernous internal carotid artery and that in the occipital vein and measured using lateral DSA ipsilateral to the BAVM—were collected. Multivariate logistic regression was performed to identify key factors associated with BAVM-related headaches.</div></div><div><h3>Results</h3><div>This study included 93 patients, among whom 50 (53.3 %) presented with headaches. The multivariate analysis revealed that female sex, occipital BAVM location, venous reflux in the superior sagittal sinus, and occipital circulation time &lt;2.37 s were independently associated with BAVM-related headache risk. We developed a model that combined angioarchitectural features and QDSA results; the model exhibited improved performance in determining BAVM-related headache risk (area under the curve: 0.854).</div></div><div><h3>Conclusions</h3><div>Female sex, occipital BAVM location, venous reflux in the superior sagittal sinus, and a short occipital circulation time are associated with headaches in patients with unruptured BAVM. QDSA can objectively quantify hemodynamic changes in patients with BAVM-related headaches.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112402"},"PeriodicalIF":3.3,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145004263","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting tumor response to TACE plus lenvatinib and PD-1 inhibitors for unresectable HCC: A multicenter observational study","authors":"Li-Wei Deng ,&nbsp;Qing-Yun Xie ,&nbsp;Bo Peng ,&nbsp;Yong Zhao ,&nbsp;Bin Liu ,&nbsp;Shi-Feng Feng ,&nbsp;Dong-Xu Liu ,&nbsp;Yan-Yuan Sun ,&nbsp;Hai-Qing Wang ,&nbsp;Chang-Li Liao ,&nbsp;Yan-Ling Wang ,&nbsp;Jun-Feng Liu ,&nbsp;Chi Zhang ,&nbsp;Yan Chen ,&nbsp;Guo-Hui Xu ,&nbsp;Le Liu ,&nbsp;Lei Cao ,&nbsp;Guo Wei ,&nbsp;Yi Ren ,&nbsp;Xue-Gang Yang","doi":"10.1016/j.ejrad.2025.112401","DOIUrl":"10.1016/j.ejrad.2025.112401","url":null,"abstract":"<div><h3>Objectives</h3><div>Preoperatively identifying patients with unresectable hepatocellular carcinoma (uHCC) who are likely to achieve an objective response to the treatment regimen of transarterial chemoembolization (TACE) plus lenvatinib and programmed death-1 inhibitors (TLP) remains challenging. We aimed to develop and validate a predictive model for tumor response to TLP treatment in patients with uHCC.</div></div><div><h3>Materials and Methods</h3><div>Patients with uHCC who received TLP treatment were divided into training (n = 107), internal validation (n = 46), and external validation (n = 52) cohorts. A nomogram model was developed based on the clinical variables of the training cohort using multivariate logistic regression. The performance of this nomogram model was evaluated using the area under the curve (AUC) and calibration curves, and its performance was compared with that of other predictive models.</div></div><div><h3>Results</h3><div>The Eastern Cooperative Oncology Group performance status, albumin–bilirubin grade, platelet-to-lymphocyte ratio, tumor distribution, and total bilirubin were identified as independent predictors of objective response. These variables were incorporated to develop the EAPTT model. The AUCs of the EAPTT model were 0.84, 0.90, and 0.85 in the training, internal validation, and external validation cohorts, respectively—statistical analysis via the DeLong test showed that these AUCs were significantly higher than those of the other seven predictive models. Stratification of patients into objective responders and non-responders via the EAPTT model revealed statistically significant progression-free survival and overall survival differences between the two groups.</div></div><div><h3>Conclusion</h3><div>The EAPTT model may enable precise stratification of the efficacy of patients with uHCC receiving TLP treatment, serving to assist in identifying the optimal candidates.</div></div>","PeriodicalId":12063,"journal":{"name":"European Journal of Radiology","volume":"192 ","pages":"Article 112401"},"PeriodicalIF":3.3,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144989221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}