European Radiology最新文献

{"title":"Prediction of thyroid malignancy risk using clinical and ultrasonography features and a machine learning approach.","authors":"Seyed Mahdi Hosseini Sarkhosh, Nooshin Shirzad, Mahdieh Taghvaei, Seyed Mohammad Tavangar, Sara Farhat, Hojat Ebrahiminik, Mahboobeh Hemmatabadi","doi":"10.1007/s00330-025-11434-2","DOIUrl":"https://doi.org/10.1007/s00330-025-11434-2","url":null,"abstract":"<p><strong>Objective: </strong>This study aims to develop and validate a predictive model for thyroid nodule malignancy risks using clinical and ultrasonography features and a machine learning (ML) approach.</p><p><strong>Methods: </strong>This retrospective study is based on the clinical and ultrasound characteristics of 1035 thyroid nodules (845 benign and 190 malignant) to develop and validate the risk prediction model. Employing multiple logistic regression, key features were selected in developing the model. Eight ML algorithms were evaluated for predicting the risks of malignancy. Finally, the predictive ability of the best-performing algorithm was compared against American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) and American Thyroid Association (ATA) guidelines.</p><p><strong>Results: </strong>Based on AUC criteria (88.3, 95% CI: 81.2-94.2), sensitivity (84.2, 95% CI: 71.1-94.7), specificity (92.3, 95% CI: 88.2-95.9), positive predictive value, (71.4, 95% CI: 60.4-83.3) and negative predictive value (96.3, 95% CI: 93.5-98.8), the XGBoost algorithm exhibited superior performance over the other ML algorithms and ACR TI-RADS and ATA. These criteria were obtained for ACR TI-RADS at 54.2%, 63.2%, 48.5%, 21.1%, and 84.8%, while for ATA, they were 44.3%, 76.3%, 27.2%, 18.4%, and 81.6%. In addition, the unnecessary fine-needle aspiration (FNA) rate with ACR TI-RADS and ATA was 43% and 63%, respectively-significantly higher than the 7% obtained with XGBoost.</p><p><strong>Conclusions: </strong>This study demonstrated the capability of ML approaches in enhancing the accuracy of predicting thyroid malignancy risks as well as their potential benefits in optimizing healthcare resources by reducing unnecessary FNA rates. Using the proposed model through a web-based tool can facilitate clinical judgments in thyroid nodule management and personalized treatment.</p><p><strong>Key points: </strong>Question Current risk assessment systems have limitations, with high unnecessary FNA rates compared to machine learning (ML) models. Findings The XGBoost algorithm was compared to other ML algorithms, ACR TI-RADS, and ATA and demonstrated superior performance. Clinical relevance This study demonstrated the capability of ML approaches in enhancing the accuracy of predicting thyroid malignancy. The proposed web-based tool to facilitate the prediction of thyroid nodule risk is available at https://aimedlab.ir/tnr .</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143413871","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CT-based detection of clinically significant portal hypertension predicts post-hepatectomy outcomes in hepatocellular carcinoma.","authors":"Subin Heo, Boryeong Jeong, Seung Soo Lee, Minju Kim, Hyeon Ji Jang, Se Jin Choi, Kang Mo Kim, Tae-Yong Ha, Dong-Hwan Jung","doi":"10.1007/s00330-025-11411-9","DOIUrl":"https://doi.org/10.1007/s00330-025-11411-9","url":null,"abstract":"<p><strong>Background: </strong>While the CT-based method of detecting clinically significant portal hypertension (CSPH) emerged as a noninvasive alternative for evaluating CSPH, its predictive ability for post-hepatectomy outcomes is unknown. Therefore, this study aimed to evaluate the impact of CT-based CSPH on outcomes following hepatectomy for hepatocellular carcinoma (HCC).</p><p><strong>Methods: </strong>This retrospective single-center study included patients with advanced chronic liver disease (ACLD) who underwent hepatectomy for very early or early-stage HCC between January 2017 and December 2018. CSPH was assessed using CT-based criteria, which included splenomegaly determined by deep learning-based spleen volume measurements with personalized reference thresholds, and the presence of gastroesophageal varices (GEV), spontaneous portosystemic shunt or ascites. Logistic regression and competing risk analyses were used to identify factors associated with severe post-hepatectomy liver failure (PHLF), hepatic decompensation, and liver-related death or transplantation. The predictive performance of existing models for PHLF was compared using both CT-based and conventional CSPH criteria (endoscopic GEV or splenomegaly with thrombocytopenia).</p><p><strong>Results: </strong>Among 593 patients (460 men; mean age 57.9 ± 9.3 years), 41 (6.9%) developed severe PHLF. The median follow-up period was 62 months. CT-based CSPH independently predicted severe PHLF (OR 7.672 [95% CI 3.209-18.346]), hepatic decompensation (subdistribution hazard ratio (sHR) 4.518 [1.868-10.929]), and liver-related death or transplantation (sHR 2.756 [1.315-5.773]). When integrated into existing models, CT-based CSPH outperformed conventional CSPH in predicting severe PHLF (AUC 0.724 vs. 0.694 for EASL algorithm (p = 0.036) and 0.854 vs. 0.830 for Wang's model (p = 0.011)).</p><p><strong>Conclusions: </strong>CT-based CSPH is a strong predictor of poor post-hepatectomy outcomes in HCC patients with ACLD, offering a noninvasive surgical risk assessment tool.</p><p><strong>Key points: </strong>Question Can CT-based detection of clinically significant portal hypertension (CSPH) serve as a noninvasive predictor of post-hepatectomy outcomes in hepatocellular carcinoma (HCC) patients? Findings CT-based CSPH independently predicted severe post-hepatectomy liver failure, hepatic decompensation, and liver-related death or transplantation, outperforming conventional CSPH criteria in predictive performance. Clinical relevance CT-based CSPH offers a noninvasive and effective tool for surgical risk assessment in HCC patients, potentially improving the selection of candidates for hepatectomy and optimizing patient outcomes.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of magnetocardiography and coronary computed tomographic angiography for detection of coronary artery stenosis and the influence of calcium.","authors":"Tingting Wu, Xin Zhao, Lanxin Feng, Shuwen Yang, Haoran Xing, Zhao Ma, Xueyao Yang, Min Zhang, Ming Ding, Yi He, Chenchen Tu, Xiantao Song, Hongjia Zhang","doi":"10.1007/s00330-025-11389-4","DOIUrl":"https://doi.org/10.1007/s00330-025-11389-4","url":null,"abstract":"<p><strong>Objectives: </strong>This study aimed to compare the diagnostic performance of magnetocardiography (MCG) and coronary computed tomography angiography (CCTA) in detecting coronary artery stenosis in relation to coronary calcification.</p><p><strong>Methods: </strong>A total of 587 patients who underwent invasive coronary angiography (ICA) with both CCTA and MCG between September 1, 2022, and August 31, 2023, were included. The patients were divided into three subgroups based on their coronary artery calcium score (CACS), namely less than 100, 100-400, and 400 and above, as determined by the Agatston score. The diagnostic sensitivity, specificity, accuracy, and the area under the receiver operating characteristic curve (ROC) of MCG, CCTA, and the combined diagnostic model (CCTA + MCG) were compared across all CACS subgroups.</p><p><strong>Results: </strong>According to ICA, 481 out of 587 patients (81.94%) had ischemia. The area under the ROC curve (AUC) of MCG for detecting ischemia was 0.80, with a sensitivity of 74.64% and specificity of 84.91% for all patients. In the different CACS subgroups, the diagnostic specificity of CCTA notably decreased (78.57% vs 24.13% vs 17.46%), while that of MCG remained stable (92.86% vs 86.21% vs 82.54%). The diagnostic accuracy of MCG and the combined diagnostic model was better than that of CCTA when CACS was ≥ 400 (77.22% vs 67.22% vs 58.89%). The AUC values of MCG, CCTA, and the combined model in the CACS ≥ 400 subgroups were 0.78, 0.49, and 0.71, respectively.</p><p><strong>Conclusions: </strong>The diagnostic performance of MCG is less affected by CACS than that of CCTA. MCG and the combined model demonstrate better performance than CCTA alone in detecting coronary artery stenosis, particularly in cases with CACS ≥ 400.</p><p><strong>Key points: </strong>Question How does the diagnostic performance of MCG compare with coronary computed tomographic angiography (CCTA) at different levels of calcification scores (CACS)? Findings MCG demonstrated better performance than CCTA in detecting coronary artery stenosis, particularly in patients with high CACS. Clinical relevance MCG or the MCG and CCTA combined model can be used to improve the noninvasive imaging diagnostic performance for detecting coronary artery stenosis and reduce unnecessary ICA, especially for patients with high calcification scores.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143424869","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"CT radiomics-based intratumoral and intertumoral heterogeneity indicators for prognosis prediction in gastric cancer patients receiving neoadjuvant chemotherapy.","authors":"Jiazheng Li, Zhenhui Li, Yinkui Wang, Yuzhuo Li, Jing Zhang, Ziyu Li, Lei Tang","doi":"10.1007/s00330-025-11430-6","DOIUrl":"https://doi.org/10.1007/s00330-025-11430-6","url":null,"abstract":"<p><strong>Objectives: </strong>CT-based intratumoral and intertumoral heterogeneity indicators were integrated to develop a prognostic model for locally advanced gastric cancer (LAGC) patients undergoing neoadjuvant chemotherapy (NACT).</p><p><strong>Methods: </strong>This retrospective study included 568 LAGC patients treated with NACT from two hospitals. The intratumor heterogeneity score (ITHscore) was developed to quantify the intratumoral heterogeneity of LAGCs on CT; intertumoral heterogeneity was characterized by combining the primary tumor (PT) and lymph node (LN) sizes on CT. CT indicators were measured on baseline and posttreatment CT scans; the reduction rates (%Δ) were calculated. The overall survival (OS) of all patients was recorded. Cox regression analysis was used to construct a preoperative survival prediction model (Pre-SPM) based on the baseline indicators and %Δ indicators. The predictive performance of Pre-SPM for OS was assessed. The clinicopathological data, including the ypTNM stage, were also collected to evaluate their impact on OS.</p><p><strong>Results: </strong>Patients with lower baseline ITHscore had better prognoses (p < 0.001). Approximately 13.01% of patients exhibited contradictory changes in PT and LN sizes. Cox regression analysis selected the baseline ITHscore, baseline PT area, %ΔPT, and %ΔLN to establish the Pre-SPM. In the external validation cohort, the c-index of Pre-SPM for predicting OS was 0.72, while the AUC for predicting 5-year OS was 0.73. After adjusting for the influence of clinicopathological features, including the ypTNM stage, Pre-SPM remained an independent prognostic factor.</p><p><strong>Conclusion: </strong>The Pre-SPM model, combining intratumoral heterogeneity and intertumoral heterogeneity, has the potential to predict the OS of LAGC patients receiving NACT.</p><p><strong>Key points: </strong>Question Increased tumor heterogeneity in LAGC affects prognosis, but effective non-invasive CT methods for assessing intratumoral and intertumoral heterogeneity are lacking. Findings ITHscore indicates intratumoral heterogeneity, while changes in PT and LN sizes reflect intertumoral heterogeneity. The Pre-SPM model, integrating both, independently predicts patient outcomes. Clinical relevance Pre-SPM enhances prognosis prediction by quantifying intratumoral and intertumoral heterogeneity, potentially guiding more personalized and effective treatment strategies for patients with LAGC.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Retrieval-augmented generation improves precision and trust of a GPT-4 model for emergency radiology diagnosis and classification: a proof-of-concept study.","authors":"Anna Fink, Johanna Nattenmüller, Stephan Rau, Alexander Rau, Hien Tran, Fabian Bamberg, Marco Reisert, Elmar Kotter, Thierno Diallo, Maximilian F Russe","doi":"10.1007/s00330-025-11445-z","DOIUrl":"https://doi.org/10.1007/s00330-025-11445-z","url":null,"abstract":"<p><strong>Objectives: </strong>This study evaluated the effect of enhancing a GPT-4 model with retrieval-augmented generation on its ability to diagnose and classify traumatic injuries based on radiology reports.</p><p><strong>Materials and methods: </strong>In this prospective proof-of-concept study, we used retrieval-augmented generation as a zero-shot learning approach to provide expert knowledge from the RadioGraphics top ten reading list for trauma radiology to the GPT-4 model, creating the context-aware TraumaCB. Radiological report findings of 50 traumatic injuries were independently generated by two radiologists. The performance of the TraumaCB compared to the generic GPT-4 was evaluated by three board-certified radiologists, assessing the accuracy and trustworthiness of the chatbot responses in the 100 reports created.</p><p><strong>Results: </strong>The TraumaCB achieved 100% correct diagnoses, 96% correct classification, and 87% correct grading, outperforming the generic GPT-4 with 93% correct diagnoses, 70% correct classification, and 48% correct grading. TraumaCB sources consistently achieved a median rating of 5.0 for explanation and trust. Challenges encountered mainly involved traumatic injuries lacking widely accepted classification systems.</p><p><strong>Conclusion: </strong>Augmenting a commercial GPT-4 model with retrieval-augmented generation improves its diagnostic and classification capabilities, positioning it as a valuable tool for efficiently assessing traumatic injuries across various anatomical regions in trauma radiology.</p><p><strong>Key points: </strong>Question Retrieval-augmented generation has the potential to enhance generic chatbots with task-specific knowledge of emergency radiology. Findings The TraumaCB excelled in accuracy, particularly in injury classification and grading, and provided explanations along with the sources used, increasing transparency and facilitating verification. Clinical relevance The TraumaCB provides accurate, fast, and transparent access to trauma radiology classifications, potentially increasing the efficiency of image interpretation in emergency departments and enabling customized reports based on local or individual preferences.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Atypical pulmonary cysts in lung cancer screening: prevalence, outcomes, and clinical implications.","authors":"Sung Hyun An, Seulgi You, You Na Kim, Joo Sung Sun","doi":"10.1007/s00330-025-11427-1","DOIUrl":"https://doi.org/10.1007/s00330-025-11427-1","url":null,"abstract":"<p><strong>Objective: </strong>We aimed to examine the prevalence and clinical outcomes of atypical pulmonary cysts (APCs) in lung cancer screening (LCS) and evaluate the impact of APC introduction on Lung-RADS v2022 categorization.</p><p><strong>Materials and methods: </strong>We reviewed 2968 low-dose baseline chest CT scans from LCS participants between June 2017 and April 2023. Two thoracic radiologists assessed the presence and type of APCs and evaluated interval changes. We also investigated how the introduction of APC changed Lung-RADS categories. Interobserver agreement was assessed using Cohen's kappa coefficient.</p><p><strong>Results: </strong>Of the 2968 participants (2900 men, 68 women; median age: 61 years [IQR: 57-65]), 18 (0.6%) had APCs. The interobserver agreement for APC presence was fair (κ = 0.33; 95% CI: 0.07-0.44). Among the 16 APC cases with follow-up CTs (median follow-up: 22 months), lung cancer was confirmed in three cases (18.8%). Among the remaining 13 APCs, 4 (30.8%) showed growth, 7 (53.8%) showed no change and 2 (15.4%) exhibited decreased wall thickness. Histological confirmation was not obtained for these 13 cases, indicating a minimum cancer rate of 18.8%. The introduction of APC led to a positive result change in 0.4% of participants (12/2968), with two lung cancer cases among them.</p><p><strong>Conclusion: </strong>APCs were infrequent in LCS, with a 0.6% prevalence, but had a high cancer rate (18.8%). Although the introduction of APC led to a 0.4% increase in positive screenings, this is considered acceptable as it included two lung cancer cases, highlighting the importance of recognizing APCs for early lung cancer detection.</p><p><strong>Key points: </strong>Question What is the prevalence and clinical significance of APCs detected during LCS, and how do they impact diagnostic outcomes? Findings APCs were detected in 0.6% of participants, with a high cancer rate of at least 18.8%, and increased screening positive results by 0.4%. Clinical relevance APCs during LCS, although uncommon, are crucial because of their high risk of lung cancer. This enhances early detection of lung cancer and improves patient management strategies.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143424828","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting progression-free survival using dynamic contrast-enhanced imaging-based radiomics in advanced nasopharyngeal carcinoma patients treated with nimotuzumab.","authors":"Xin Jin, Wen-Zhu Li, Yi-Hao Guo, Gang Wu, Wei-Yuan Huang, Feng Chen","doi":"10.1007/s00330-025-11433-3","DOIUrl":"https://doi.org/10.1007/s00330-025-11433-3","url":null,"abstract":"<p><strong>Purpose: </strong>The purpose of this study was to explore the potential value of the radiomics model based on dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), compared with the clinical model mostly based on the epidermal growth factor receptor (EGFR) expression, in predicting progression-free survival (PFS) in patients with locally advanced nasopharyngeal carcinoma (LA-NPC) treated with nimotuzumab (NTZ).</p><p><strong>Methods: </strong>A total of 136 patients with LA-NPC who received NTZ treatment between January 2018 and June 2022 were included in this study. Patients were randomly divided into training (n = 95) and validation (n = 41) groups in a 7:3 ratio. DCE-MRI radiomics, clinical, and clinical-radiomics models were built to predict PFS. The relationship between EGFR expression levels and NTZ efficacy was assessed using Kaplan-Meier curves. Model performance was assessed using the area under the curve, calibration, and DeLong tests. Decision curve analysis evaluated net clinical benefit. Patients were stratified into high- and low-risk groups based on optimal model radiomic scores, and prognoses were compared using Kaplan-Meier curves.</p><p><strong>Results: </strong>Univariate Cox regression analysis demonstrated that EGFR expression level was the only independent predictive factor of PFS (p < 0.05). Patients with EGFR 3+ receiving NTZ therapy had significantly longer PFS than those with EGFR 1+ (hazard ratio = 3.025, p < 0.05). The clinical-radiomics model exhibited superior predictive efficacy, compared with the radiomics and clinical models (training set: 0.887 vs. 0.845, 0.654; validation set: 0.831, 0.824 vs. 0.567, all p < 0.001).</p><p><strong>Conclusions: </strong>The clinical-radiomics models using DCE-MRI and EGFR levels can effectively predict NTZ efficacy in LA-NPC patients, providing objective evidence for personalized treatment adjustments.</p><p><strong>Key points: </strong>Question How can the response to nimotuzumab treatment in patients with locally advanced nasopharyngeal carcinoma be accurately predicted using non-invasive imaging methods? Findings A combined clinical and radiomic model using dynamic contrast-enhanced magnetic resonance imaging showed improved predictive performance for progression-free survival in patients treated with nimotuzumab. Clinical relevance The study provides evidence for using a combined clinical and radiomic approach, offering a non-invasive method to predict treatment response and guide personalized treatment strategies for patients with locally advanced nasopharyngeal carcinoma, potentially improving patient outcomes.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143425274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multicenter diagnostic study of thyroid nodule with Hashimoto's thyroiditis enabled by Hashimoto's thyroiditis nodule-artificial intelligence model.","authors":"Chen Chen, Yahan Zhou, Bo Xu, Lingyan Zhou, Mei Song, Shengxing Yuan, Wenwen Yue, Yibo Zhou, Hangjun Chen, Ruyi Yan, Benlong Xiao, Tian Jiang, Qi Zhang, Shanshan Zhao, Changsong Xu, Chenke Xu, Jiao Lu, Lin Sui, Yuqi Yan, Mingshun Lyu, Qingquan He, Vicky Yang Wang, Dong Xu","doi":"10.1007/s00330-025-11422-6","DOIUrl":"https://doi.org/10.1007/s00330-025-11422-6","url":null,"abstract":"<p><strong>Objective: </strong>This study aimed to develop a Hashimoto's thyroiditis nodule-artificial intelligence (HTN-AI) model to optimize the diagnosis of thyroid nodules with Hashimoto's thyroiditis (HT) of which the efficiency and accuracy remain challenging.</p><p><strong>Design and methods: </strong>This study included 5709 patients from 10 hospitals between January 2014 and March 2024. Among them, 5053 thyroid nodules were divided into training and testing sets in a 9:1 ratio. Then, we tested the model on an external dataset (n = 432). Finally, we prospectively recruited 224 patients with dynamic ultrasound videos acquired and employed the HTN-AI model to identify nodules from the dynamic ultrasound videos. Radiologists of varying seniority performed the categorization of thyroid nodules as benign and malignant, both with and without the assistance of the HTN-AI model, and their diagnostic performances were compared.</p><p><strong>Results: </strong>The results indicated that for the external testing set, the HTN-AI model achieved a Dice similarity coefficient (DSC) of 0.91, outperforming several other common convolutional neural network (CNN) models. Specifically, the DSCs of the HTN-AI model were similar for thyroid nodule patients with and without HT which were 0.91 ± 0.06 and 0.91 ± 0.09. Moreover, when the HTN-AI model was used to assist diagnosis, it demonstrated an improvement in the diagnostic performance of radiologists. The diagnostic areas under the receiver operating characteristic curve (AUCs) of the junior radiologists increased from 0.59, 0.59, and 0.57 to 0.68, 0.65, and 0.65.</p><p><strong>Conclusions: </strong>This research demonstrates that the HTN-AI model has excellent performance in identifying thyroid nodules associated with HT and can assist radiologists with more accurate and efficient diagnoses of thyroid nodules.</p><p><strong>Key points: </strong>Question The study developed an HTN-AI model aimed at assisting in the diagnosis of thyroid nodules in patients with HT. Findings The HTN-AI model achieved great performance with a Dice similarity coefficient (DSC) of 0.91, and consistent performance across patients with and without HT. Clinical relevance The HTN-AI model enhances the accuracy and efficiency of thyroid nodule diagnosis, particularly in patients with HT. By assisting radiologists at varying experience levels, this model supports improved decision-making in the management of thyroid nodules.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143406408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comprehensive morphomechanical and wall enhancement analysis of intracranial aneurysms.","authors":"Sebastian Sanchez, Jacob M Miller, Adam Galloy, Marco Nino, Diego Ojeda, Elena Sagues, Carlos Dier, Andres Gudino, Rodrigo Jaramillo, Edgar A Samaniego","doi":"10.1007/s00330-025-11417-3","DOIUrl":"https://doi.org/10.1007/s00330-025-11417-3","url":null,"abstract":"<p><strong>Introduction: </strong>Understanding the relationship between aneurysm wall enhancement (AWE), wall shear stress (WSS), and wall tension (WT) across different brain aneurysm compartments may improve the assessment of rupture risk and vessel wall dynamics. Our comprehensive analysis aimed to evaluate these characteristics across different brain aneurysm compartments.</p><p><strong>Methods: </strong>Patients with intracranial aneurysms underwent 3-T high-resolution magnetic resonance imaging (MRI). T1-weighted images were obtained before and after administering gadolinium (Gd) contrast. 3D segmentations of the aneurysms were generated using 3D Slicer. AWE was quantified based on the signal intensity of the aneurysm wall on T1+Gd images. Finite element analysis calculated WT, while computational fluid dynamics assessed WSS metrics, including time-averaged wall shear stress (TAWSS), wall shear stress gradient (WSSG), and oscillatory shear index (OSI). AWE, WT, and WSS metrics were compared among aneurysms compartments and aneurysms stratified by size, aspect ratio (AR), and size ratio (SR).</p><p><strong>Results: </strong>AWE was elevated in aneurysms with greater size and AR. WT and OSI were higher in aneurysms with larger sizes and higher AR and SR. Aneurysm necks exhibited higher WT, TAWSS, and WSSG compared to the aneurysm body. Irregular aneurysms displayed lower TAWSS, WSSG, and higher OSI in areas of high AWE. The bleb had the highest AWE and lowest WT and WSSG, while necks showed the lowest AWE and highest WT and WSSG.</p><p><strong>Conclusion: </strong>Aneurysm necks typically exhibit high WT and WSS, while the body often shows low WSS. Blebs express elevated AWE alongside diminished WSS and WT which potentially increases rupture risk KEY POINTS: Question Areas of high wall enhancement in irregular aneurysms are characterized by low wall sheer stress and high oscillatory sheer index. Findings High risk structures such as daughter sacs exhibit the highest wall enhancement and lowest wall sheer stress. Clinical relevance Comprehensive morphomechanical and wall enhancement analysis of brain aneurysms have the potential to offer valuable new insights into aneurysm biology.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143406411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MRI of pediatric ovarian masses: validation of the O-RADS framework.","authors":"Alice M Munari, Caterina B Monti, Camilla Viglio, Gianluca Folco, Francesco Rizzetto, Salvatore Zirpoli","doi":"10.1007/s00330-025-11444-0","DOIUrl":"https://doi.org/10.1007/s00330-025-11444-0","url":null,"abstract":"<p><strong>Objective: </strong>The purpose of our study was to test the applicability and implications of using the O-RADS system, which is developed and validated on adults, to review MRI of ovarian masses among pediatric patients.</p><p><strong>Materials and methods: </strong>We retrospectively reviewed consecutive MRI examinations from pediatric patients referred to imaging for suspected ovarian lesions, assessing them using the O-RADS framework. Malignancy frequencies among O-RADS classes were reviewed, and we appraised the potential for such approach to split patients into low (O-RADS 1, 2, and 3) and high risk (O-RADS 4 and 5). Multivariate analyses were conducted to review which clinical or imaging variables yielded a significant impact on malignancy, and a simplified reading framework was proposed accordingly.</p><p><strong>Results: </strong>109 female patients were included, with a median age of 13 years (IQR 11-15 years), 7 (7%) presenting with malignant lesions. Malignancy proportions were 0% (95% confidence Interval (CI) 0-35%) for the O-RADS 1 class, 0% (95% CI 0 - 5%) for the O-RADS 2 class, 0% (95% CI 0-14%) for the O-RADS 3 class, 50 (95% CI 1 - 99%) for the O-RADS 4 class, and 75% (95% CI 41-93%) for the O-RADS 5 class. The presence of peritoneal thickening or nodules (p < 0.001), lesion composition (p < 0.001), and absence of intralesional fat (p = 0.051) were individual predictors of malignancy, and the simplified reading framework proposed with such variables identified 11 likely malignant cases, detecting all 7 malignant lesions.</p><p><strong>Conclusion: </strong>The O-RADS system may be applied to MRI performed in the pediatric population for ovarian masses, and a simplified reading framework based on O-RADS could also prove useful in such a setting.</p><p><strong>Key points: </strong>Question The Ovarian-Adnexal Reporting and Data System (O-RADS) provides the risk of malignancy of ovarian masses among adults but has not been validated among pediatric patients. Findings Malignancy frequencies for O-RADS classes 1, 2, 3, and 4 were 0, 0, 50%, and 75%, indicating a good accuracy in lesion discrimination. Clinical relevance The Ovarian-Adnexal Reporting and Data System (O-RADS) can be effectively applied to MRI examinations in pediatric patients, enabling accurate classification of findings, with potential for score simplification in this age group.</p>","PeriodicalId":12076,"journal":{"name":"European Radiology","volume":" ","pages":""},"PeriodicalIF":4.7,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143406444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}