Insights into Imaging最新文献

{"title":"Systemic inflammatory factors combined with HR-VWI-based plaque vulnerability discriminate ischemic phenotypes and predict prognosis in patients with intracranial atherosclerosis.","authors":"Ke Lv, Huiying Wang, Ying Liu, Xunxiao Zhao, Shuxin Ma, Wencan Fu, Yue Cheng, Shuang Xia","doi":"10.1186/s13244-026-02395-1","DOIUrl":"10.1186/s13244-026-02395-1","url":null,"abstract":"<p><strong>Background: </strong>The relationship between systemic inflammation, plaque vulnerability, and stroke phenotypes in intracranial atherosclerotic stenosis (ICAS) is unclear. This study investigated inflammatory markers' association with high-risk plaques and their value in differentiating stroke phenotypes and predicting prognosis.</p><p><strong>Materials and methods: </strong>298 symptomatic ICAS patients were classified into acute ischemic stroke (AIS, n = 207) and delayed perfusion (DP, n = 91) groups using multimodal MRI. Plaque vulnerability was graded based on high-resolution vessel wall imaging, inflammatory indices calculated, and associations assessed using regression and mediation analysis.</p><p><strong>Results: </strong>AIS patients had higher systemic inflammation and more high-risk plaques (all p < 0.05). Each Systemic Inflammation Response Index (SIRI) unit increase was associated with 32.4% higher AIS risk (p = 0.027) and nearly threefold increased 90-day poor outcome risk (p < 0.001). In an exploratory mediation analysis aimed at understanding how these markers jointly differentiate the two phenotypes, the association between SIRI and AIS was partially (25.9%) statistically accounted for by high-risk plaque features. A comprehensive model integrating clinical, inflammatory, and imaging markers effectively differentiated AIS from DP (AUC = 0.794). In an exploratory prognostic analysis with a small number of poor outcomes, adding SIRI improved the AUC from 0.610 to 0.959, with internal validation supporting robustness (mean cross-validated AUC: 0.919-0.939), p < 0.05. Associations were more pronounced in males.</p><p><strong>Conclusions: </strong>Systemic inflammation is associated with ischemic phenotype in ICAS, and this association may be partially explained by plaque vulnerability. Combined assessment effectively differentiates stroke phenotypes and predicts prognosis, particularly in males, supporting multi-parametric models for early identification of at-risk patients.</p><p><strong>Key points: </strong>Question: Can systemic inflammatory markers combined with high-risk plaque features discriminate acute ischemic stroke from delayed perfusion and predict prognosis in patients with intracranial atherosclerotic stenosis?</p><p><strong>Findings: </strong>Systemic inflammation response index was independently associated with acute ischemic phenotypes and poor prognosis. Multi-parametric models improved phenotype discrimination (AUC = 0.794) and prognosis prediction (cross-validated AUC: 0.919-0.939). Critical relevance statement: Systemic inflammation response index is associated with acute ischemic stroke in intracranial atherosclerotic stenosis, an association partially mediated by high-risk plaque features. Combined assessment of inflammatory and imaging biomarkers improves stroke phenotyping and prognosis prediction, particularly in male patients.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13546255/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148897126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From zero-shot to fine-tuning: optimize large language models for error detection of ultrasound reports.","authors":"Min Lai, Jin Zhang, Yaer Lv, Chunjie Hou, Ceng Wang, Fengzhi Li, Songcheng Xu, Meixia Du, Min Wei, Dong Xu, Litao Sun","doi":"10.1186/s13244-026-02394-2","DOIUrl":"10.1186/s13244-026-02394-2","url":null,"abstract":"<p><strong>Background: </strong>High workload and inconsistent quality of ultrasound report writing may lead to diagnostic errors. This study aims to ascertain whether fine-tuned open-source large language models (LLMs) can achieve promising performance for automated quality control of Chinese ultrasound reports, when compared to proprietary LLMs.</p><p><strong>Materials and methods: </strong>This retrospective, multi-center study included a multi-subspecialty dataset of 1800 Chinese ultrasound reports, comprising 1500 quality-controlled reports injected artificially with six predefined error types and 300 reports with naturally occurring errors. Nine proprietary LLMs (under zero-shot and few-shot paradigms) and seven open-source LLMs (under fine-tuning) were evaluated, with performance compared against that of radiologists of varying seniority. Performance was measured by detection accuracy, Macro-F1 score, precision, recall, and mean absolute error across six categories.</p><p><strong>Results: </strong>Fine-tuned open-source LLMs, notably Qwen3-14B, achieved a detection accuracy of 0.931 and a Macro-F1 of 0.739, approaching the performance of senior radiologists. Some fine-tuned open-source LLMs maintained performance despite smaller parameter sizes and outperformed most proprietary LLMs with vastly larger parameter counts. The fine-tuned Qwen3-14B demonstrated superior recognition capability for semantic errors such as redundancy, spelling, orientation, and unit or value errors.</p><p><strong>Conclusion: </strong>This study demonstrates that task-specific fine-tuning enables open-source LLMs to rival proprietary LLMs and expert radiologists in Chinese ultrasound report error detection, offering a locally deployable and privacy-compliant alternative for AI-assisted clinical quality control workflows.</p><p><strong>Key points: </strong>Question Can task-specific fine-tuning improve error detection by open-source LLMs in Chinese ultrasound reports and provide an effective approach to automated report quality control? Findings Task-specific fine-tuning enabled open-source LLMs to achieve Macro-F1 scores up to 0.739, approaching that of experienced radiologists (0.764) and outperforming most proprietary LLMs. Critical relevance statement Task-specific fine-tuning allows open-source LLMs to become feasible assistants in ultrasound report quality control workflows, offering a locally deployable, privacy-preserving, and regulation-compliant solution for enhancing reporting consistency and reducing diagnostic errors in high-volume ultrasound examination procedures.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13546399/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148897374","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"How do non-radiologist physicians read and use the radiology report? A French national survey.","authors":"Yasmine Kassab, Matthieu Bailly, Sixtine Brabant, Catherine Oppenheim, Alice Bernard-Tessier, Natacha Stolowy, Helene Salaün, Maelle Dejobert, Thibaut Lemoine, Marine Begon, Grégoire Boulouis, Jean-Philippe Cottier, Marie-Agnes Lauvin, Helena Daurel, Anne-Laure Piton, Raoul Kanav Khanna, Gilles Metrard, Clara Cohen","doi":"10.1186/s13244-026-02387-1","DOIUrl":"10.1186/s13244-026-02387-1","url":null,"abstract":"<p><strong>Objectives: </strong>We developed a national questionnaire to assess radiology reports (RR) physicians' reading habits and identify their expectations towards RR, focusing on CT and MRI reports.</p><p><strong>Materials and methods: </strong>An anonymized 20-item questionnaire was nationally distributed in France via practice and hospital mailing lists and one social media group, and included three sections: respondent's profile; RR reading (importance and reading of RR sections, advice on key images [i.e., selected summary screenshots for main findings]); use and impact of the RR. Overall and profile-based subgroup analyses were performed.</p><p><strong>Results: </strong>One thousand two hundred fifty physicians responded (58% women, 71% aged 20-34 years, 53.4% juniors, 85% from medical specialties). RR reading: 60.3% frequently read the whole report, 17.4% conclusion-only, and 9.9% images-only. Most respondents considered key images essential for comprehension, positively influencing opinion on RR and reinforcing confidence in the conclusion; 63.3% agree with a systematic key-images inclusion within RR. Concerning preliminary reports, 49.3% await validation by a senior radiologist to make their decision. RR use and impact: 55.7% copy the radiologist's conclusion into their record, 39.5% reformulate it. Seven percent admitted having made a medical error by not reading the RR carefully. In life-threatening emergencies, 75% review the RR retrospectively, 18.5% refer directly to specialists regardless of the radiologist's opinion. Responses were influenced by respondents' profile (medical specialists vs surgeons, academic status, experience, gender), especially regarding the importance of the radiologist's name and the incorporation of the conclusion.</p><p><strong>Conclusion: </strong>Most clinicians declared reading the entire RR. Key images add value, while clinician profile shapes the way they read and use RR.</p><p><strong>Key points: </strong>Question: How do clinicians read and incorporate RR into medical records, and does the clinician's profile influence these practices?</p><p><strong>Findings: </strong>Out ot 1253 respondents, 60.3% read the entire report or the findings section depending on the conclusion; 55.7% copy the radiologist's conclusion. Critical relevance statement: RR's conclusions are commonly incorporated as is into the record. Careful conclusion formulation, together with the addition of illustrating key images, may facilitate report use. These findings prompt consideration of tailoring reporting practices to the profile of the requesting physician.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13546390/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148896781","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Noninvasive differentiation of benign and malignant solid pulmonary nodules using multiparameter dual-layer spectral CT radiomics.","authors":"Yanyan Wang, Ye Yu, Yicheng Fu, Ying Zhang, Yu Wang, Xiao Yu, Baocong Liu, Yan Zhou, Huawei Wu","doi":"10.1186/s13244-026-02380-8","DOIUrl":"10.1186/s13244-026-02380-8","url":null,"abstract":"<p><strong>Objectives: </strong>To establish and validate a noninvasive multiparametric radiomics model based on dual-layer spectral CT (DLCT) for distinguishing preoperatively malignant from benign solid pulmonary nodules (SPNs).</p><p><strong>Materials and methods: </strong>This retrospective study enrolled 441 patients with pathologically confirmed SPNs who underwent preoperative DLCT and were divided into training (n = 252), internal test (n = 112), and external test (n = 77) cohorts. Radiomics features were extracted from conventional and virtual monoenergetic images (40 and 70 keV) and material decomposition images (including iodine density (ID), Z-effective atomic number (Zeff), and electron density (ED) maps) in arterial (AP) and venous (VP) phases. Logistic regression was used to construct radiomics models, and a combined clinical-radiomics model was visualized as a nomogram. Subgroup analysis was performed by nodule size (≤ 10 mm vs.  > 10 mm), and diagnostic accuracy was compared with that of two radiologists.</p><p><strong>Results: </strong>The optimal radiomics model, comprising features from ID in VP and Zeff in both AP and VP, achieved an area under the curve (AUC) of 0.835, 0.804, and 0.772 in the training, internal, and external cohorts, respectively. The combined model (age, lobulation, and ten radiomic features) outperformed the clinical-radiological model in all cohorts (AUC: 0.889 vs. 0.816; 0.865 vs. 0.795; 0.825 vs. 0.743; p < 0.05). It maintained strong performance for ≤ 10 mm and > 10 mm nodules, with AUCs of 0.875 and 0.888 (internal test) and 1.000 and 0.794 (external test).</p><p><strong>Conclusion: </strong>A DLCT-based multiparametric radiomics model, integrated with clinical-radiological features, enables accurate preoperative, noninvasive differentiation between benign and malignant SPNs, including subcentimeter nodules.</p><p><strong>Key points: </strong>Question: Accurate preoperative differentiation of solid pulmonary nodules remains clinically challenging because benign and malignant nodules often show overlapping features on conventional CT.</p><p><strong>Findings: </strong>A combined model integrating multiparameter dual-layer spectral CT radiomics and clinical-radiological features outperformed the clinical-radiological model in differentiating benign and malignant solid pulmonary nodules. Critical relevance statement: The developed model provides an accurate, noninvasive tool for preoperative differentiation of indeterminate solid pulmonary nodules, supporting clinical management decisions.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13545195/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148891472","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"High-pitch free-breathing triple rule-out CT angiography for acute chest pain: a prospective sequential two-cohort comparison of radiation dose and diagnostic image quality.","authors":"Juan Yu, Zhoufeng Peng, Xi Yu, Benyang Wang, Wenbing Yang, Wenyang Pan, Yunfei Zha","doi":"10.1186/s13244-026-02390-6","DOIUrl":"10.1186/s13244-026-02390-6","url":null,"abstract":"<p><strong>Objectives: </strong>To evaluate the feasibility of high-pitch free-breathing triple rule-out CT angiography (TRO-CTA) by comparing image quality and radiation dose with a standard breath-hold protocol in patients with acute chest pain.</p><p><strong>Materials and methods: </strong>This prospective study enrolled 104 patients with acute chest pain who underwent TRO-CTA on a third-generation dual-source CT scanner, sequentially allocated to high-pitch free-breathing TRO-CTA (pitch = 3.2; Group A, n = 52) or standard breath-hold TRO-CTA (standard pitch; Group B, n = 52). Objective image quality (attenuation, noise, signal-to-noise ratio [SNR], and contrast-to-noise ratio [CNR]) was assessed across the coronary, pulmonary, and aortic territories. Subjective image quality (3-point scale), effective dose (ED), acquisition time, and clinical diagnoses were compared between groups.</p><p><strong>Results: </strong>Radiation dose (ED: 1.81 ± 1.33 vs 15.63 ± 6.71 mSv) and acquisition time (0.48 ± 0.05 vs 6.83 ± 1.87 s) were significantly lower in Group A than in Group B (all p < 0.001). Vascular attenuation was significantly higher in Group A in the coronary arteries, main pulmonary artery, and ascending aorta (p < 0.05), whereas CNR was higher in Group B (all p < 0.01). Subjective image quality scores, including coronary diagnostic segment rates (98.0% vs 97.5%), were comparable across all vascular territories (all p > 0.05). Clinical diagnoses were similarly distributed between groups (p = 0.455).</p><p><strong>Conclusion: </strong>High-pitch free-breathing TRO-CTA substantially reduces radiation dose while preserving comparable diagnostic image quality, offering a clinically feasible low-dose alternative to standard breath-hold protocols for emergency patients with acute chest pain, particularly those with well-controlled heart rates.</p><p><strong>Key points: </strong>Question: Conventional TRO-CTA involves high radiation exposure (8.93-17.5 mSv) and breath-holding, posing safety and feasibility challenges in emergency patients with acute chest pain.</p><p><strong>Findings: </strong>The high-pitch free-breathing protocol reduced effective dose by 88% and acquisition time from 6.83 to 0.48 s while providing comparable diagnostic image quality. Critical relevance statement: High-pitch free-breathing TRO-CTA may offer a low-dose imaging option on dual-source CT, with the potential to support more time-efficient emergency evaluation of acute chest pain.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13545161/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148891526","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MST-Net: a multi-scale spatial transformation network for automated T2/T3 staging of rectal cancer on preoperative T2-weighted MRI.","authors":"Wei Liu, Ling Yin, Xiong Ran, Tian-Xiang Xu, Yun Du, Jiang-Yi He, Shao-Quan Zhou, Jian-Jun Li","doi":"10.1186/s13244-026-02377-3","DOIUrl":"10.1186/s13244-026-02377-3","url":null,"abstract":"<p><strong>Objective: </strong>This study aimed to develop a deep learning model (MST-Net) for the segmentation-free automated and precise differentiation between T2 and T3 staging of rectal cancer based on preoperative T2-weighted magnetic resonance imaging (T2WI).</p><p><strong>Materials and methods: </strong>Preoperative T2WI images and postoperative pathologically confirmed T2/T3 staging labels from rectal cancer patients were retrospectively collected from two centers between January 2020 and January 2025. The core of MST-Net includes: a pyramid cross-stage feature fusion module to integrate shallow texture details and deep semantic information, mitigating the loss of spatial details in deep networks; and a lightweight spatial attention mechanism to adaptively focus on key radiomics features within the tumor region. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and accuracy (ACC).</p><p><strong>Results: </strong>A total of 309 patients were included in the analysis, comprising a training set (n = 216) and an internal test set (n = 54) from Center 1, and an external independent validation set (n = 39) from Center 2. MST-Net achieved an AUC of 0.95 (ACC = 90%) on the internal test set, outperforming junior radiologists (ACC = 69%) and performing comparably to senior radiologists (ACC = 85%). On the external validation set, it achieved an AUC of 0.85 (ACC = 79%), also surpassing junior radiologists (ACC = 65%) and reaching the level of senior radiologists (ACC = 72%).</p><p><strong>Conclusion: </strong>MST-Net enables segmentation-free automated and high-accuracy prediction of T2/T3 staging in rectal cancer based on routine T2WI images. This model provides a reproducible auxiliary tool for clinical personalized treatment decision-making, potentially reducing the dependence of staging results on physician experience.</p><p><strong>Key points: </strong>Question Accurate differentiation between T2 and T3 rectal cancer on preoperative T2-weighted MRI remains clinically challenging but is essential for treatment planning. Findings MST-Net achieved high diagnostic accuracy and robust discrimination for T2 versus T3 staging in both internal and external validation cohorts. Critical Relevance This study demonstrates that a segmentation-free deep learning model using routine T2-weighted MRI can enable accurate preoperative T-staging of rectal cancer, supporting standardized and efficient radiological assessment and facilitating more consistent clinical decision-making in rectal cancer management.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13545167/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148891534","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From fluid loss to calcification: MRI and CT matrix features in cystic echinococcosis.","authors":"Marine Julien, Louis Bohard, Andréas Busse-Coté, Noémie Tissot, Celia Turco, Coralie Barerra, Laurence Millon, Solange Bresson-Hadni, Paul Calame","doi":"10.1186/s13244-026-02375-5","DOIUrl":"10.1186/s13244-026-02375-5","url":null,"abstract":"<p><strong>Objective: </strong>To characterize MRI matrix signal and CT calcification patterns across WHO-IWGE stages, compare them with lesion mimicking cystic echinococcosis (CE), and derive an MRI-based subclassification for inactive cysts.</p><p><strong>Materials and methods: </strong>This retrospective multicenter study included 105 patients (174 CE cysts) imaged between 2012 and 2024, plus 30 mimickers. Inclusion required axial T1-weighted imaging with chemical-shift in-phase/opposed-phase acquisitions and at least two T2-weighted sequences. Two radiologists independently reviewed all examinations, and a consensus was reached. Cysts were classified as inactive if no free-fluid T2 hypersignal was present on index MRI and corroborated by follow-up imaging and/or adjunct ultrasound in equivocal cases. Diagnostic performance indices were calculated to identify inactivity.</p><p><strong>Results: </strong>Fatty T1 hyperintensity was more frequent in inactive than active/transitional cysts (20/97 (21%) vs 2/77 (3%); p < 0.001). The \"matrix flip-flop\" sign was observed in 61/97 (63%) inactive cysts and in only two CE3b lesions (2/77 (3%); p < 0.001). Both signs showed high specificity for inactivity (75/77; 97.4%, 95% CI: 90.9-99.7). Matrix calcifications were associated with inactivity (46/97 (47%) vs 2/77 (3%); p < 0.001) and were rare among mimickers (2/30 (7%); p = 0.006). Homogeneous non-fatty T1 hyperintensity characterized mimickers (23/30 (77%) vs 10/174 (6%); p < 0.001). A three-tier inactive subclassification (early/intermediate/late) was derived from matrix T2 evolution.</p><p><strong>Conclusion: </strong>Matrix MRI/CT features-particularly fatty foci and the matrix flip-flop sign-provide ancillary evidence of CE inactivity and help distinguish CE from mimickers. T2 signal supported an MRI-based framework for stage-adapted \"watch-and-wait\" follow-up.</p><p><strong>Key points: </strong>Question MRI features of cystic echinococcosis remain insufficiently studied, particularly matrix signal abnormalities that may affect diagnosis and viability assessment on routine imaging. Findings T1- and T2-weighted matrix CE cysts analysis identified fatty foci and the newly described matrix flip-flop sign as specific features of inactive cysts. Critical Relevance Assessment of underrecognized MRI and CT cyst-matrix features can help stage cystic echinococcosis, assess viability, and distinguish cystic echinococcosis cysts from mimickers.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13542020/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148887491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Inter-reader agreement in multimodality imaging of carotid plaque-RADS: considerations for modality selection.","authors":"Jie Zhang, Yuzhu Yan, Yingqi Xing, Liyang Bai, Rina Sa, Lijuan Wang","doi":"10.1186/s13244-026-02393-3","DOIUrl":"10.1186/s13244-026-02393-3","url":null,"abstract":"<p><strong>Objectives: </strong>Plaque-RADS was recently developed to standardize reporting and risk stratification of carotid plaques. This study aimed to evaluate inter-reader agreement of Plaque-RADS across US, CTA, and MRI, and to assess the impact of reader experience.</p><p><strong>Materials and methods: </strong>In 250 patients with carotid US, CTA, and MRI, nine readers (three per modality, stratified by experience) independently assessed plaque features and Plaque-RADS scores. Inter-reader agreement was analyzed using κ statistics and intraclass correlation coefficients (ICC). Learning curves were generated by comparing reader performance against MRI expert reference across five sequential case subsets.</p><p><strong>Results: </strong>Among 962 plaques and 25 normal arteries, inter-modality agreement for overall Plaque-RADS was moderate to substantial (κ, 0.578-0.758). Agreement was almost perfect for RADS 1-2 (κ, 0.814-0.883) but lower for RADS 3-4 (κ, 0.427-0.709). US, CTA, and MRI demonstrated substantial to almost perfect agreement in assessing plaque presence and maximum wall thickness (κ/ICC, 0.697-0.878). CTA and MRI experts demonstrated substantial to almost perfect agreement in ulcer detection (κ, 0.880) and intraluminal thrombus (κ, 0.787), but moderate agreement in intraplaque hemorrhage (κ, 0.578). Regarding reader experience, all intermediate and expert readers across US, CTA, and MRI achieved substantial to almost perfect agreement in overall and different categories of RADS scoring (κ/ICC, 0.624-0.954). Learning curve analysis confirmed that all readers improved with training, with the most pronounced gains observed in US beginners.</p><p><strong>Conclusion: </strong>Plaque-RADS is reliable for low-risk plaques across modalities, whereas accurate categorization of high‑risk plaques requires experienced readers and advanced imaging. These findings may help inform stratified imaging decisions.</p><p><strong>Key points: </strong>Question Does inter‑reader agreement for the Carotid Plaque‑RADS score differ across ultrasound, CTA, and MRI, and how does reader experience affect consistency? Findings Plaque‑RADS showed excellent agreement for low‑risk plaques across all modalities. High‑risk plaques required advanced imaging and experienced readers; structured training improved performance, especially for ultrasound. Critical Relevance Statement This study systematically compares Plaque‑RADS agreement across ultrasound, CTA, and MRI. The findings support a stratified imaging strategy (US for low‑risk assessment, MRI for high‑risk plaque characterization) pending prospective validation.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13538297/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148880027","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Delphi consensus of specialist radiologists on imaging criteria for gadolinium contrast use and whole-body MRI in prostate cancer.","authors":"Eric Nduka Onwuharine, Alexander James Clark, Matthew Dimmock, Jonathan Hill","doi":"10.1186/s13244-026-02388-0","DOIUrl":"10.1186/s13244-026-02388-0","url":null,"abstract":"<p><strong>Objective: </strong>To achieve expert consensus on structured prostate MRI criteria for gadolinium-based contrast agent (GBCA) administration and whole-body MRI (WB-MRI) escalation in the diagnostic evaluation and staging of prostate cancer (PCa).</p><p><strong>Materials and methods: </strong>A three-round modified online Delphi survey was conducted among UK specialist prostate radiologists. Participants rated predefined clinical and MRI criteria using structured Likert agreement scales and provided free-text comments to refine decision-making. Quantitative responses were summarised descriptively. Qualitative thematic analysis informed subsequent rounds. Consensus for each decision criterion was predefined as ≥ 70% agreement. The study was conducted in accordance with CREDES (Conducting and Reporting of Delphi Studies) guidance.</p><p><strong>Results: </strong>The number of radiologists who completed Rounds 1, 2, and 3 was 21, 14, and 11, respectively. Two principal outputs were defined: (1) consensus criteria for GBCA administration; (2) consensus criteria for WB-MRI escalation. For GBCA administration, consensus supported its use for Prostate Imaging-Reporting and Data System (PI-RADS) 3 lesions (86%) and low prostate imaging quality (PI-QUAL ≤ 3; 91%), and avoidance in high-quality (PI-QUAL 4-5) images (73%). Fixed prostate-specific antigen (PSA) or PSA density thresholds did not reach consensus when considered in isolation. For WB-MRI, consensus supported escalation for T3a/T3b disease (93%), combined clinical and radiological suspicion of metastasis (91%), and high-risk profiles (PSA ≥ 20 ng/mL, Gleason ≥ 8 or ≥ T3; 86%).</p><p><strong>Conclusion: </strong>This Delphi study established expert consensus on structured MRI criteria to guide selective GBCA administration and WB‑MRI escalation, with the potential to reduce unnecessary GBCA administration, improve consistency and optimise prostate MRI workflow.</p><p><strong>Key points: </strong>Question What consensus-based criteria can guide on-table gadolinium-based contrast agent administration and whole-body MRI escalation during prostate MRI to reduce variability and unnecessary imaging? Findings Delphi consensus established imaging-based criteria integrating lesion suspicion, image quality and high-risk features to guide selective gadolinium-based contrast agent administration and whole-body MRI escalation. Critical relevance statement This article critically evaluates gadolinium-based contrast agent administration in prostate MRI and defines consensus-based standardised decision-making criteria. These criteria aim to reduce unnecessary gadolinium-based contrast agent administration and improve efficiency and consistency in staging pathways.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13534386/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148873948","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ANATOMIADI: a national, image-based anatomy competition for residents in Italy to enhance postgraduate education through interactive learning.","authors":"Carlo Cosimo Quattrocchi, Marinella Neri, Monica Antolini, Alessio Apollonio, Andrea Casula, Flora De Rosa, Veronica Domestico, Marianna Fusina, Paolo Giorgini, Giuseppe Greco, Daniel Raimi, Fabrizio Sannino, Daniele Vo, Andrea Falzone","doi":"10.1186/s13244-026-02389-z","DOIUrl":"10.1186/s13244-026-02389-z","url":null,"abstract":"<p><strong>Background: </strong>For radiology residents, the steep learning curve of cross-sectional anatomy (especially in CT and MRI) demands more efficient and engaging educational tools. In recent years, game-based learning (GBL) and gamification have emerged as powerful strategies to enhance postgraduate education by leveraging 'active recall' and 'spaced repetition' within a competitive framework. Here we describe the context of ANATOMIADI and assess the perceived educational impact.</p><p><strong>Materials and methods: </strong>Image datasets were anonymized, archived, and manually labeled to highlight specific anatomical structures. Over 400 unique games were developed, categorized by anatomical system and tiered into three levels of difficulty. A structured feedback survey was designed and distributed to the participants after the event to capture data on different domains regarding the educational value, gamification and engagement, platform usability and team collaboration, technical workflow, logistics, scientific accuracy, fairness, and quality assessment.</p><p><strong>Results: </strong>ANATOMIADI challenge registered an attendance of 141 participants representing 28 residency programs across Italy. The evaluation of the event's perceived value yielded a unanimous positive response; indeed, 100% of the participants recommended that the competition be repeated in future academic years. Based on the required level of knowledge, the event was recommended to residents in their 3rd (54/63 responses) and 4th year (48/63 responses) of postgraduate education.</p><p><strong>Conclusion: </strong>By integrating high-fidelity imaging with interactive game mechanics, the ANATOMIADI challenge not only fostered anatomical learning but also built the foundation for a more collaborative and technologically adept generation of radiologists.</p><p><strong>Key points: </strong>Question: In the era of high-resolution multidetector CT and advanced MRI, the volume of anatomical data that a resident in radiology must process is unprecedented.</p><p><strong>Findings: </strong>We developed a context using image-based puzzle games to challenge different residency schools in radiology on imaging-based anatomical structures and rules. Critical relevance statement: ANATOMIADI is the first national competition based on radiological anatomy and dedicated to residents. By using an inclusive \"twinning\" framework among residency programs, it transforms dense theoretical study into a highly motivating, collaborative, and social learning experience.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-08-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13525974/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148857423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}