Insights into Imaging最新文献

{"title":"Letter to the Editor: Closing the loop in radiology-patient access to reports as an additional safety layer.","authors":"Yoshito Tsushima, Ayako Taketomi-Takahashi","doi":"10.1186/s13244-026-02412-3","DOIUrl":"10.1186/s13244-026-02412-3","url":null,"abstract":"","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13598025/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149024739","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":"Commercial large language models for oral cavity cancer staging using descriptive pre-treatment MRI reports: ready for standalone use in clinical practice?","authors":"Qi Yong Hemis Ai, Hoi Ming Kwok, Ming-Yi Lu, Tracy T S Lau, Kuo Feng Hung, Lun M Wong, Tiffany Y So, Ann D King, Ho Sang Leung","doi":"10.1186/s13244-026-02409-y","DOIUrl":"10.1186/s13244-026-02409-y","url":null,"abstract":"<p><strong>Objectives: </strong>MRI is used for staging head and neck cancer (HNC), but assigning T- and N-category criteria requires specialised expertise, and so many institutions offer only descriptive MRI reports. This study assessed potentials of commercial large language models (LLMs) to stage oral cavity cancer (OCC) using descriptive MRI reports and compared their accuracy with human experts.</p><p><strong>Materials and methods: </strong>104 eligible MRI reports were processed by five commercial LLMs (ChatGPT5.4, ChatGPT5.0, ChatGPT4.1, Gemini3.1 and DeepSeekV3.2). T- and N-categories, and overall stage were extracted from outputs of the LLMs. MRI reports were also staged by two multidisciplinary team (MDT) members. Accuracies of the LLMs and MDT members for cancer staging were assessed against the reference staging standard and compared using the McNemar test.</p><p><strong>Results: </strong>The LLMs showed accuracy of 54.8-76.0% for T-categorisation, 63.5-88.5% for N-categorisation, and 53.8-80.8% for overall stage. Compared with DeepSeekV3.2, ChatGPT and Gemini3.1 showed significantly higher accuracy (p ≤ 0.001), except for Gemini3.1 for T-categorisation (p = 0.07). No differences in accuracy for staging between ChatGPT versions and between them and Gemini3.1 (p = 0.08 to > 0.99). MDT members showed accuracy of 74.0-76.9% for T-categorisation, 76.9-77.9% for N-categorisation and 68.3-71.2% for overall stage. The tested LLMs did not consistently outperform MDT members for staging.</p><p><strong>Conclusion: </strong>Variable tested LLM performance in staging OCC from descriptive MRI reports suggested that they are not suitable as standalone staging tools. Additionally, Low MDT staging accuracy highlighted the need for structured reports that include clinical cancer staging to facilitate MDT assessment, thus potentially ensuring optimised disease management.</p><p><strong>Key points: </strong>Question Can commercial LLMs accurately assign cancer stage based on descriptive MRI reports to overcome the lack of specialised expertise in clinical practice? Findings Tested LLMs demonstrated variable performances, ranging 54.8-76.0% for T-categorisation, 63.5-88.5% for N-categorisation and 53.8-80.8% for overall stage, and failed to consistently outperform MDT members. Critical relevance statement Variable tested LLM performance in staging OCC from descriptive MRI reports suggested that they are not suitable as standalone staging tools. MDT members' low performances highlight that structured MRI reports with clinical cancer staging are needed to improve multidisciplinary assessment.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13594034/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149016709","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":"MRI insights into endometrial stromal tumors.","authors":"Le Wang, Qi-Wen Hu, Yu Zou","doi":"10.1186/s13244-026-02400-7","DOIUrl":"10.1186/s13244-026-02400-7","url":null,"abstract":"<p><p>Preoperative misdiagnosis and the adoption of incorrect surgical approaches occur frequently in patients with endometrial stromal tumors (ESTs) because of their rarity, nonspecific clinical manifestations, and imaging features that highly overlap with those of other benign uterine tumors, thereby increasing the risk of secondary surgery and adverse outcomes. Thus, accurate diagnosis is essential, and following initial clinical and sonographic assessments, MRI serves as the cornerstone imaging technique for further characterization and differential diagnosis. Radiologists must be familiar with and adept at interpreting these images to narrow the differential diagnosis and at least suggest the possibility of EST. This article describes the imaging manifestations of various ESTs and highlights characteristic radiological features that enable radiologists to refine the differential diagnosis, thereby assisting clinicians in determining appropriate treatment strategies. KEY POINTS: Question Preoperative differentiation of ESTs from benign uterine lesions remains challenging and may result in the adoption of the incorrect surgical approach. Findings Key MRI features include homogeneous T2WI hyperintensity without speckles, T2WI hypointense bands, a worm‑like appearance, multilocular cystic changes, and feather‑like enhancement. Critical Relevance Statement This review comprehensively describes the MRI findings of ESTs and provides an educational summary based on tumor location, morphology, and key features to substantially improve diagnostic confidence.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13594005/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149016674","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":"Post-therapy imaging of adult brain tumours: patterns, pitfalls, and practical framework.","authors":"Mina F G Isaac","doi":"10.1186/s13244-026-02398-y","DOIUrl":"10.1186/s13244-026-02398-y","url":null,"abstract":"<p><p>Post-therapy imaging of adult brain tumours is among the most challenging tasks in neuroradiology. Distinguishing true tumour progression from treatment-related changes-including pseudoprogression, radiation necrosis, and atypical immunotherapy responses-requires integration of imaging timing, morphology, and clinical context. This review synthesises current literature into a practical, treatment-based framework, addressing gliomas and metastases separately and, within each, organising findings by treatment modality-surgical resection, chemoradiotherapy, stereotactic radiosurgery, systemic therapy (targeted and anti-angiogenic agents, and immunotherapy), and locoregional techniques. For each treatment context, we outline expected imaging evolution, characteristic patterns of treatment effect, key features differentiating these from tumour progression, and the role of MRI and nuclear medicine techniques. The review concludes with a practical reporting framework built around treatment received and time since treatment, integrating clinical context, imaging pattern, and interval change to support radiology reporting across diverse therapeutic combinations. CRITICAL RELEVANCE STATEMENT: Recognising treatment-specific imaging pitfalls-pseudoprogression, pseudoresponse, radiation injury, and immune-related changes-is essential for accurate response assessment and prevents misdirected management in neuro-oncology. KEY POINTS: Distinguishing true tumour progression from treatment-related change on post-therapy brain MRI is a frequent diagnostic challenge, because findings vary with treatment type and timing. Pseudoprogression is context-dependent-its definition, timing, and morphology differ across chemoradiotherapy, IDH-mutant glioma, stereotactic radiosurgery, and immunotherapy, so no single rule fits all settings.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13589733/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149008273","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":"Intracranial venous thrombosis imaging by anatomical location: an educational review with mimics and diagnostic pitfalls.","authors":"Anna Falk Delgado, David Fällmar, Heather Martin, Michael V Mazya, Fabian Arnberg, Annika Kits","doi":"10.1186/s13244-026-02381-7","DOIUrl":"10.1186/s13244-026-02381-7","url":null,"abstract":"<p><p>Intracranial venous thrombosis encompasses diverse etiologies with clinical presentations ranging from an incidental finding to coma and death. Despite potentially devastating consequences, venous thrombosis is frequently overlooked by radiologists due to its predominantly extraaxial location close to the skull and clinical emphasis on arterial imaging evaluation in acute stroke management. This educational review addresses common and uncommon presentations across all anatomical locations of intracranial venous thrombosis, providing practical pearls and pitfalls for diagnostic interpretation. We review the imaging modalities CT and MRI, together with their venographic techniques, correlating imaging findings with anatomical location, thrombus age and signal evolution, complications, and underlying causes. Critical interpretive pitfalls are highlighted with strategies to avoid misdiagnosis. The review also emphasizes the important yet underrecognized association between intracranial pressure disorders and venous thrombosis, providing radiologists with a practical framework for improved diagnostic accuracy. KEY POINTS: Question How can radiologists systematically recognize intracranial venous thrombosis across all anatomical locations and avoid the diagnostic pitfalls that lead to clinically significant missed diagnoses? Findings Location-specific imaging patterns, thrombus signal evolution, and recognition of common mimics such as arachnoid granulations enable accurate detection using combined non-contrast and venographic sequences. Critical relevance statement By organizing intracranial venous thrombosis imaging anatomically and pairing each location with its specific mimics and pitfalls, this review equips radiologists with a practical framework to improve detection accuracy and prevent clinically significant missed diagnoses.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13588991/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149001732","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":"Gynecologic imaging errors and pitfalls: how to avoid misdiagnosis on pelvic MRI.","authors":"Alexandre Faure, Edwige Pottier, Marie Florin, Léo Razakamanantsoa, Pascal Rousset, Isabelle Thomassin-Naggara","doi":"10.1186/s13244-026-02378-2","DOIUrl":"10.1186/s13244-026-02378-2","url":null,"abstract":"<p><p>Pelvic magnetic resonance imaging (MRI) plays a pivotal role in evaluating gynecologic diseases, offering superior soft-tissue characterization, multiplanar imaging, and functional assessment compared to ultrasound and computed tomography. This article proposes evidence-based, standardized MRI protocols and a reproducible interpretative framework to enhance diagnostic accuracy, reduce misdiagnoses, and support multidisciplinary management in routine clinical practice. Key insights cover MR imaging of major gynecologic diseases: endometriosis, adnexal masses including the O-RADS MR scoring system, and pathological hypothesis as well as uterine diseases including endometrial, cervical, and myometrial benign and malignant conditions. For all situations, we present typical and atypical presentations as well as mimickers. This article will give some tips and tricks to help the radiologist in clinical routine for detecting and differentiating the main issues that impact patient management. By adopting standardized protocols and reports with adequate pathology knowledge, radiologists can improve patient outcomes through precise staging, surgical planning, and personalized care. KEY POINTS: Question How can radiologists avoid common diagnostic errors and pitfalls when interpreting pelvic MRI for gynecologic diseases, including endometriosis, uterine tumors, and adnexal masses? Findings Knowledge of common pitfalls, standardized MRI protocols with Dixon-based T1-weighted sequences, diffusion-weighted imaging, and dynamic contrast enhancement, if necessary, combined with structured reporting frameworks including O-RADS MRI and FIGO 2023, reduces misdiagnosis across gynecologic conditions. Critical Relevance Statement This article critically reviews typical and atypical presentations alongside common mimickers of major gynecologic diseases on pelvic MRI, providing radiologists with practical evidence-based strategies to improve diagnostic accuracy, reduce unnecessary interventions, and support multidisciplinary clinical decision-making.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13586104/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148989896","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":"Planning and performing CT-guided interventions for source control in septic patients: a survey by the European Society of Emergency Radiology.","authors":"Anne Frisch, Maria Isabel Opper Hernando, Kerstin Rubarth, Ann-Christine Stahl, Ana Blanco Barrio, Raffaella Basilico, Vittorio Miele, Francesca Iacobellis, Myrto Bolanaki, Denis Witham, Marc Dewey, Julian Pohlan","doi":"10.1186/s13244-026-02374-6","DOIUrl":"10.1186/s13244-026-02374-6","url":null,"abstract":"<p><strong>Objectives: </strong>To analyze emergency radiologists' perspectives on computed tomography (CT)-guided interventions for source control in patients with sepsis.</p><p><strong>Materials and methods: </strong>In 2023, members of the European Society of Emergency Radiology (ESER) were surveyed on the use of CT in the management of sepsis (n = 297). The questionnaire included demographic details, clinical experience, and questions on the timing and justification for CT. In a secondary analysis focused on CT-guided interventions, the obtained data were compared to results from a previous single-center survey among clinicians and general radiologists (n = 719). Responses were gathered anonymously, and data were analyzed using descriptive statistics and chi-square tests.</p><p><strong>Results: </strong>The perspectives of 144 emergency radiologists were compared with those of 719 participants from various specialties. Most respondents, i.e., 85.3% (n = 58) of emergency radiologists, strongly agreed on the importance of CT for planning interventions in patients with sepsis. Preferences for the timing of CT-guided intervention leaned towards 1-6 h among emergency radiologists (66.7%, n = 36) and clinicians (64.1%, n = 211), while a majority of general radiologists accepted 6-12-h (58.3%, n = 14). Professional experience did not affect the agreement on the importance of CT among emergency radiologists (board-certified, 85.7% (n = 30) vs senior, 84.8% (n = 28)) or timing preferences (board-certified, 60.7% (n = 17) vs senior, 73.1% (n = 19)).</p><p><strong>Conclusion: </strong>The ESER survey confirmed the importance physicians attach to CT for planning interventions in patients with sepsis. While preferences for intervention timing varied slightly among specialties, there was a wide agreement on the importance of prompt CT-guided interventions within the first 12 h of focus detection.</p><p><strong>Key points: </strong>Question What is the expert assessment on the significance of CT imaging for procedure planning and on the ideal timing of CT-guided interventions in sepsis. Findings the surveyed experts value CT for planning interventions in patients with sepsis, and a prompt CT-guided intervention ideally within the 6 h of focus detection. Critical relevance statement specialized imaging techniques and minimally invasive treatments for source control become more widely used. The perspective of emergency radiologists compared to that of general radiologists and clinicians helps to weigh the role of CT-guided interventions in septic patients.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13578189/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148976864","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":"Correction: Development and validation of fat-corrected virtual MR elastography to assess fibrosis stage in metabolic dysfunction-associated steatotic liver disease.","authors":"Jie Yuan, Xinxin Xu, Fuhua Yan, Zhiwei Qin, Huamei Yan, Wenjie Yang, Huimin Lin","doi":"10.1186/s13244-026-02369-3","DOIUrl":"10.1186/s13244-026-02369-3","url":null,"abstract":"","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13562469/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148929852","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":"Letter to the Editor: Cerebral CT perfusion dose: diagnostic reference levels should stimulate, not replace, optimization.","authors":"Thomas Stein, Alexander Rau, Elias Kellner","doi":"10.1186/s13244-026-02379-1","DOIUrl":"10.1186/s13244-026-02379-1","url":null,"abstract":"","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13553993/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148911614","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":"Evaluating the biopsy-negative prostate lesions based on sequential biparametric prostate magnetic resonance imaging: Can PRECISE be used to guide repeat biopsy?","authors":"Yuxuan Tian, Mingjian Ruan, Yuke Chen, Qi Shen, Shuoyan Chen, Jingyun Wu, Derun Li, Yu Fan, Yi Liu","doi":"10.1186/s13244-026-02373-7","DOIUrl":"10.1186/s13244-026-02373-7","url":null,"abstract":"<p><strong>Objective: </strong>Indications for repeat biopsy after an initial negative prostate biopsy remain poorly defined. Therefore, this study aimed to investigate the effectiveness of the Prostate Cancer Radiological Estimation of Change in Sequential Evaluation (PRECISE) scoring system as a guide for repeat biopsy.</p><p><strong>Materials and methods: </strong>We retrospectively reviewed patients who underwent repeat prostate biopsy and had bpMRI performed before both biopsies between 2013 and 2024. The association of clinically significant prostate cancer (csPCa) detection with initial Prostate Imaging Reporting and Data System (PI-RADS) and PRECISE scores was analyzed. The predictive effects of prostate-specific antigen density before repeat biopsy (re-PSAD) and PSA velocity (PSAV) on csPCa were assessed in patients progressing to or maintaining re-PI-RADS 3.</p><p><strong>Results: </strong>104 individuals were included with a median follow-up period of 28.58 months and a PSAV of 0.90 ng/mL/year (interquartile range (IQR), -0.17 to 2.57 ng/mL/year). The csPCa detection rate at repeat biopsy was 24.0%. 42 participants (40.4%) had PRECISE scores of 4-5, 43 (41.3%) had scores of 3, and 19 (18.3%) had scores of 1-2. A higher PRECISE score corresponds to a higher csPCa detection rate, with csPCa diagnosed in 52.4% of patients with PRECISE scores of 4-5. In subgroup analyses, re-PSAD and PSAV avoided 65.4% and 53.8% of repeat biopsies, respectively, and would have caused 5.9% and 0.00% of missed csPCa.</p><p><strong>Conclusions: </strong>PRECISE may be useful for guiding repeat biopsy decisions and may provide incremental value for predicting csPCa when interpreted together with PI-RADS. Additionally, PSAD and PSAV can assist in screening for repeat biopsy.</p><p><strong>Key points: </strong>Question Indications for repeat prostate biopsy after an initial negative result remain poorly defined. Can sequential MRI PRECISE scoring help inform repeat biopsy decisions? Findings In our retrospective cohort, PRECISE 1-2 yielded no clinically significant cancer, while PRECISE 4-5 showed a 52.4% detection rate. PSA parameters aided equivocal cases. Critical relevance statement This study evaluates sequential MRI PRECISE scoring, suggesting its potential to help reduce unnecessary repeat biopsies for regressing lesions while identifying high-risk progressing abnormalities.</p>","PeriodicalId":13639,"journal":{"name":"Insights into Imaging","volume":"17 1","pages":""},"PeriodicalIF":5.6,"publicationDate":"2026-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13550309/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148906102","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}