BMC Medical Imaging最新文献

{"title":"Effectiveness of artificial intelligence-assisted colonoscopy for colorectal lesion detection: a systematic review and meta-analysis of randomized controlled trials.","authors":"Xia Liu, Jian Qin, Yuxia Zhao, Chen Zhang, Dengchao Wang, Tao Jiang","doi":"10.1186/s12880-026-02586-8","DOIUrl":"10.1186/s12880-026-02586-8","url":null,"abstract":"<p><strong>Background: </strong>Colonoscopy is the gold standard for colorectal cancer detection and prevention, but its performance depends on operator skill and lesion features. Artificial intelligence (AI)-assisted colonoscopy has been introduced to improve adenoma detection.</p><p><strong>Methods: </strong>A comprehensive search was conducted in PubMed, Embase, the Cochrane Library, Web of Science, ClinicalTrials.gov, and major Chinese databases to identify randomized controlled trials (RCTs) comparing AI-assisted and conventional colonoscopy. Meta-analyses were performed using Review Manager 5.4 and Stata 14.0. The quality of evidence was graded using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system, while trial sequential analysis (TSA) was applied to assess the robustness and information sufficiency of the results.</p><p><strong>Results: </strong>A total of 46 RCTs encompassing 37,012 participants were included. Compared with conventional colonoscopy, AI-assisted colonoscopy was associated with a higher ADR (RR = 1.22, 95% CI = 1.17-1.27, P < 0.00001), polyp detection rate (PDR: RR = 1.22, 95% CI = 1.16-1.29, P < 0.00001), polyp detection per colonoscopy (PPC: MD = 0.28, 95% CI = 0.21-0.36), and adenoma detection per colonoscopy (APC: MD = 0.22, 95% CI = 0.15-0.29). The detection rate of sessile serrated lesions (SSL) was also higher in the AI-assisted group (RR = 1.20, 95% CI: 1.05-1.36, P = 0.007), whereas no significant difference was observed in advanced adenoma detection rate (AADR).</p><p><strong>Conclusions: </strong>AI-assisted colonoscopy was associated with improvements in several detection-related quality indicators, including ADR, PDR, PPC, APC, and SSL detection rate. However, given the very low GRADE certainty for ADR and PDR and the lack of long-term outcome data, these findings should be interpreted cautiously.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13692893/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149242760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MRI-based radiomics models for the preoperative prediction of massive intraoperative blood loss in spinal metastases patients.","authors":"Heng Deng, Jiayang Yan, Jiayi Zhang, Fukai Li, Guangwen Duan, Wenhao Jia, Xiang Wang, Shiyuan Liu","doi":"10.1186/s12880-026-02541-7","DOIUrl":"10.1186/s12880-026-02541-7","url":null,"abstract":"<p><strong>Objectives: </strong>To develop and validate MRI-based radiomics models for predicting intraoperative massive blood loss (MBL) in patients with spinal metastases.</p><p><strong>Materials and methods: </strong>A total of 507 patients diagnosed with spinal metastases were enrolled in this study, who were classified as MBL and non-MBL group, with the 2500 ml as the threshold. Radiomic features were extracted from T2WI and CET1 sequences and dimensionality reduction was performed by LASSO regression analysis. Radiomics models were developed using radiomics features, yielding a radiomics signature from the best model. Clinical variables were analyzed to create a clinical model. The combined model incorporated both clinical variables and Rad-signature. The predictive performance was assessed through the area under the curve (AUC), accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and F1 score. Calibration curves and decision curve analyses (DCA) were generated to evaluate the model's accuracy and clinical utility. Finally, a nomogram was developed to visualize the optimal model.</p><p><strong>Results: </strong>Tumor vascularity, preoperative embolization, tumor location, and surgical segments were independent risk factors for MBL. Among 11 machine learning models evaluated, the AdaBoost model demonstrated optimal predictive performance (AUC = 0.775) and was selected as the radiomics model. The combined radiomics-clinical model achieved the highest discriminative ability (AUC = 0.818; 95% CI: 0.7191-0.9165), outperforming both the radiomics model (AUC = 0.768) and the clinical model (AUC = 0.583). The nomogram showed good calibration and provided the greatest net clinical benefit.</p><p><strong>Conclusions: </strong>Radiomics-based prediction models enable preoperative assessment of MBL risk in surgery and support personalized perioperative management strategies.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13700267/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149262362","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Value of delayed post-contrast phase hyperintensity on gadobenate dimeglumine-enhanced MRI for differentiating atypical splenic hemangiomas from cysts: an exploratory retrospective study with pathology validation.","authors":"Meina Li, Kaixian Lin, Feifei Yu, Yijing Zhao, Meilian Xiong, Dairong Cao, Zhen Xing","doi":"10.1186/s12880-026-02615-6","DOIUrl":"10.1186/s12880-026-02615-6","url":null,"abstract":"<p><strong>Background: </strong>Incidental splenic cystic lesions are commonly encountered on MRI. Atypical hemangiomas with absent enhancement on conventional MRI (cMRI) may be misdiagnosed as simple cysts, yet they carry a small risk of spontaneous rupture. Delayed post‑contrast phase imaging using gadobenate dimeglumine (Gd-BOPTA) may detect contrast retention in hemangiomas, offering a potential discriminator. This study provides an exploratory evaluation of whether delayed post-contrast hyperintensity can help differentiate atypical splenic hemangiomas from true cysts.</p><p><strong>Methods: </strong>In this exploratory retrospective, single‑center study, we searched our PACS database for reports of splenic cysts or hemangiomas between June 2015 and June 2025. A consecutive series of 220 patients (162 lesions initially classified as cysts on cMRI, 58 as hemangiomas) was included. Two independent radiologists, blinded to clinical data and reference standard, assessed cMRI features and delayed post-contrast phase signal intensity. The index test was delayed post-contrast phase hyperintensity (yes/no). The reference standard was histopathology in an exploratory validation subset (n = 26); for the full cohort (n = 220), imaging findings were descriptively reported. Diagnostic accuracy metrics (sensitivity, specificity, positive predictive value [PPV], negative predictive value [NPV]) were explored in the pathologically confirmed subset with 95% confidence intervals using the Wilson score method.</p><p><strong>Results: </strong>Interobserver agreement was excellent for cMRI classification (κ = 0.86) and delayed post-contrast phase signal assessment (κ = 0.88). In the full cohort (n = 220), 82/162 (50.6%) of cMRI‑diagnosed cysts exhibited delayed post‑contrast hyperintensity, representing an imaging feature discordant with a simple cyst. Among the 26 pathologically confirmed lesions (validation subset), 18 were hemangiomas and 8 were cysts. Delayed post-contrast hyperintensity demonstrated, sensitivity 94.4% (95% CI: 72.7-99.9%), specificity 100% (95% CI: 67.6-100%), PPV 100% (95% CI: 81.6-100%), and NPV 88.9% (95% CI: 56.5-98.0%). The wide confidence intervals reflect the small sample size (n = 26) and inherent imprecision of these exploratory estimates.</p><p><strong>Conclusion: </strong>Delayed post-contrast hyperintensity on Gd-BOPTA-enhanced MRI is a potential imaging sign for differentiating atypical splenic hemangiomas from true cysts. This exploratory observation warrants prospective validation before clinical adoption.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13700433/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149262360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Substantia nigra iron-related alterations, white matter disruption, and functional network dysconnectivity in temporal lobe epilepsy: a multimodal neuroimaging study.","authors":"Xu Zhao, Huiting Wu, Si Jian, Shanshan Huang, Yingying Zhou, Tingxin Yang, Yuanhao Li, Tong Tong, Zhiqiang Zhou, Suiqiang Zhu, Xiaoyan Liu, Wenzhen Zhu","doi":"10.1186/s12880-026-02613-8","DOIUrl":"10.1186/s12880-026-02613-8","url":null,"abstract":"<p><strong>Background: </strong>Although the substantia nigra (SN) has been implicated in temporal lobe epilepsy (TLE) pathophysiology, clinical studies remain scarce. This study aimed to characterize alterations in SN susceptibility, white matter (WM) microstructure, and functional connectivity (FC), as well as their associations with cognitive functions in TLE patients.</p><p><strong>Methods: </strong>This study included 20 right-TLE (RTLE), 21 left-TLE (LTLE), and 23 healthy controls (HCs). Multimodal neuroimaging was used to assess the susceptibility, diffusion tensor imaging metrics, and FC of SN subregions. Group differences were examined by analysis of covariance. Exploratory correlation analyses were performed to evaluate associations between imaging metrics and cognitive performance.</p><p><strong>Results: </strong>RTLE patients showed higher susceptibility (post-hoc P = 0.015 vs. HCs; P = 0.002 vs. LTLE) and radial diffusivity (RD) (post-hoc P = 0.014 vs. HCs; P = 0.001 vs. LTLE) in the contralateral substantia nigra pars reticulata (SNr), alongside lower fractional anisotropy (FA) (post-hoc P = 0.009 vs. HCs; P = 0.006 vs. LTLE) in the contralateral substantia nigra pars compacta (SNc). LTLE patients exhibited decreased FA in the ipsilateral SNr (post-hoc P < 0.001 vs. HCs; P = 0.001 vs. RTLE) and reduced connectivity between the left SNc and the right superior parietal lobule (P = 0.026, cluster-level FDR corrected). Susceptibility and RD positively correlated in the contralateral SNr of RTLE cohorts (r = 0.564, uncorrected P = 0.010). Exploratory analyses suggested associations between SNr susceptibility, FA, and cognitive performance in TLE patients (all uncorrected P < 0.05).</p><p><strong>Conclusions: </strong>Our multimodal findings suggest subregion- and laterality-specific alterations in the SN of patients with TLE, with patterns differing between RTLE and LTLE.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13700436/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149262350","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnosis of tympanic effusions using hyperspectral imaging to prevent unnecessary paracentesis procedures.","authors":"Maximilian Gänzle, Sabine Löffler, Denis Gholami Bajestani, Michael Unger, Andreas Dietz, Andreas Melzer, Markus Pirlich, Hannes Köhler","doi":"10.1186/s12880-026-02611-w","DOIUrl":"10.1186/s12880-026-02611-w","url":null,"abstract":"<p><strong>Background: </strong>Otitis Media with Effusion (OME) is a prevalent childhood condition, affecting up to 90% of children with 10% progressing to a chronic state. It is commonly treated by tympanostomy tube placement. Diagnosis relies on clinical history, otoscopy, audiometry, and tympanometry, but these methods have a high false-positive rate of 15-28%. This highlights the need for an objective, reliable, and non-invasive diagnostic tool to reduce unnecessary surgeries.</p><p><strong>Methods: </strong>To evaluate the rate of false-positive OME diagnoses in our tertiary hospital, a retrospective database analysis was conducted. To assess the feasibility of advanced imaging, experimental studies were performed ex vivo on human body donors. Sodium chloride solution was intratympanically injected under endoscopic guidance. Hyperspectral imaging (HSI) datasets were acquired using a customized endoscopic system to quantify tissue water content.</p><p><strong>Results: </strong>An analysis of hospital records identified a false positive rate of 29% (n = 4,564) for OME diagnosis, comparing the number of paracenteses to paracenteses plus tympanostomy tubes. In the ex vivo analysis of HSI, visual evaluation of the Tissue Water Index (TWI) demonstrated sensitivity, specificity, and accuracy of 89%, 78%, and 83%, respectively. The mean TWI of all evaluated tympanic membranes was 50.2 (± 7.1) before injection (control) and increased to 62.8 (± 14.8) after injection.</p><p><strong>Conclusion: </strong>HSI presents a highly sensitive and user-friendly potential future diagnostic adjunct for diagnosing OME, which has shown its potential in an ex vivo approach. Now it needs in-vivo confirmation of the results as its implementation in clinical routine could contribute to avoiding unnecessary surgical interventions, minimizing patient burden and healthcare costs.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13422311/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148618030","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"EPRA U-Net: an efficient pyramid residual attention framework for accurate infarct segmentation in diffusion-weighted MRI.","authors":"Hasan Ulutas, Muhammet Emin Sahin, Mustafa Fatih Erkoc, Esra Yuce, Turker Tuncer, Sengul Dogan, Serkan Kiranyaz","doi":"10.1186/s12880-026-02569-9","DOIUrl":"10.1186/s12880-026-02569-9","url":null,"abstract":"<p><strong>Objective: </strong>Accurate identification of acute ischemic infarcts on diffusion-weighted magnetic resonance imaging (DWI) is a critical prerequisite for reliable lesion quantification and effective clinical decision support in the management of cerebrovascular events.</p><p><strong>Methods: </strong>This study presents EPRA U-Net (Efficient Pyramid Residual Attention U-Net), a task-specific integrated architecture for efficient and accurate infarct segmentation of DWI images. In the proposed architecture, an EfficientNet-based encoder was used as a hierarchical feature extractor with a minimized parameterization. In addition, a Residual-Recurrent (R2) block (recurrent unrolling step t = 2, following the original formulation) and Atrous Spatial Pyramid Pooling (ASPP) were integrated to enhance the performance of spatial dependency modeling. Additionally, a dual attention mechanism was incorporated to highlight lesion-related activations while concurrently enabling the suppression of extraneous background responses. To prioritize lesion detection consistent with clinical imperative, a Tversky loss function (α = 0.4, β = 0.6) was adopted, thereby emphasizing the sensitivity of detection over its specificity during the optimization process.</p><p><strong>Results: </strong>Experimental evaluations were conducted utilizing an in-house dataset comprising 167 patients with 4,895 DWI slices; subsequently, the performance of the proposed EPRA U-Net was assessed in comparison with state-of-the-art models, specifically UNet++, DeepLabV3+, and TransUNet. The experimental results suggest that EPRA U-Net attained superior performance, evidenced by a pixel-aggregated Dice of 0.8984, a per-sample Dice of 0.9469, an IoU of 0.8155, a Recall of 0.8887, a Lesion F1 of 0.9378, and an HD95 of 11.62 px (median: 2.00 px). Furthermore, a clear reduction in the rate of missed lesions, specifically by 16%, 25%, and 29%, was observed when compared with UNet++, DeepLabV3+, and TransUNet, respectively. Additionally, a comprehensive ablation study was systematically conducted.</p><p><strong>Conclusion: </strong>These findings suggest that the proposed architecture affords a more robust and accurate infarct segmentation, consequently facilitating a more reliable estimation of treatment eligibility.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13688968/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149242791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quantitative assessment of aortic wall thickening disease by dual energy computed tomography angiography.","authors":"Na Li, Jia Liu, Qinyue Luo, Lijie Zhang, Kailu Zhang, Yu Feng, Heshui Shi","doi":"10.1186/s12880-026-02605-8","DOIUrl":"10.1186/s12880-026-02605-8","url":null,"abstract":"<p><strong>Objectives: </strong>A broad spectrum of etiologies can lead to aortic wall thickening. This study aimed to evaluate the utility of quantitative parameters derived from dual-layer computed tomography (CT) in differentiating intramural hematoma (IMH), non-calcified plaques, and aortic thrombosis, and to explore their associations with laboratory inflammatory markers.</p><p><strong>Methods: </strong>365 patients were retrospectively enrolled, including atherosclerosis (n = 98), aortic dissection (n = 126), and IMH (n = 141). Laboratory indicators were collected from the electronic medical record system. Regions of interest (ROI) of non-calcified plaques, thrombi, hematomas, and periaortic adipose tissue around lesions were outlined. The lesion enhancement values and degree of enhancement were calculated from the CT values in polyenergetic and virtual non-contrast image. The slope of the energy spectrum curve was determined using the formula: K = (CT_40keV - CT_100keV) / 60. Normalization was performed for the effective atomic number (Z), iodine density (ID), and iodine non-water density (IW) using the aortic lumen as the standard reference, resulting in normalized Z values, normalized iodine density (NID), and normalized iodine non-water density (NIW).</p><p><strong>Results: </strong>The enhancement values and degrees of enhancement differed among the three lesions (all P < 0.05). Differences in Z value, ID, IW, and K were observed among non-calcified plaques, thrombi, and hematomas (all P < 0.05). The NID of hematoma was the largest, while the NIW of thrombus was the smallest. Normalized Z value of non-calcified plaque was significantly smaller than those of thrombus and hematoma (both P < 0.05), with no difference between thrombus and hematoma (P > 0.05). The Z_{periaortic fat} pre- and post-normalization were significantly smaller around the plaques than around the thrombi and hematomas (both P < 0.05). Significant correlations were found between aortic wall normalized Z_lesion, Z_{periaortic fat}, and laboratory parameters, including neutrophil % and neutrophil count, lymphocyte % and lymphocyte count, C-reactive protein, and D-dimer.</p><p><strong>Conclusions: </strong>Quantitative spectral CT parameters show promise for the non-invasive identification of aortic wall lesions. The weak correlations with laboratory inflammatory markers suggest a potential but limited link between imaging parameters and local inflammation. These exploratory findings require further validation. If confirmed in future studies, this approach may complement conventional CTA and guide targeted therapy.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13700438/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149262309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of nasal septal deviation, age, and sex on anterior, middle, and posterior clinoid process pneumatization: a morphological computed tomography study.","authors":"Rukiye Soyal, Betül Digilli Ayaş, Aynur Emine Çiçekcibaşı, Demet Aydoğdu","doi":"10.1186/s12880-026-02610-x","DOIUrl":"10.1186/s12880-026-02610-x","url":null,"abstract":"<p><strong>Background: </strong>Clinoid process pneumatization is a critical anatomical variation that influences surgical exposure and complication risks in skull base approaches. This study aimed to comprehensively evaluate the pneumatization patterns of the anterior (ACP), middle (MCP), and posterior clinoid processes (PCP) and to determine the effects of age, sex, and nasal septal deviation (NSD) parameters, specifically angle severity and laterality, on these aeration patterns.</p><p><strong>Methods: </strong>Computed tomography images of 250 adults (125 females, 125 males; aged 18-70 years) were retrospectively evaluated. Pneumatization patterns of the ACP and PCP were classified using a four-type grading system (Type 0-III). In addition, the presence of the MCP and its pneumatization status were evaluated. NSD was evaluated for laterality and angle severity (mild:<9°, moderate:9°-15°, severe:>15°). Statistical associations were assessed using chi-square tests and multivariate regression models.</p><p><strong>Results: </strong>Type 0 (no pneumatization) predominated across all processes (ACP:81.8%, MCP:82.4%, PCP:82.6%), while Type III (total pneumatization) was rare (ACP:2.0%, PCP:2.2%). While univariate analysis showed age-related associations across all regions, multivariate logistic regression identified an independent association between age and ACP pneumatization, with younger individuals (18-34 years) exhibiting lower degrees of aeration (p < 0.01). Male sex was independently associated with bilateral MCP pneumatization (OR:3.016, p < 0.001) and right PCP pneumatization (OR:2.234, p = 0.032). Severe NSD angle (> 15°) was strongly associated with bilateral PCP pneumatization (p < 0.001) and higher-grade ACP aeration (p < 0.01). Furthermore, MCP presence showed strong ipsilateral clustering with ACP and PCP pneumatization patterns (p < 0.001).</p><p><strong>Conclusion: </strong>Clinoid process pneumatization demonstrates distinct anatomical associations with NSD angle severity, age, and sex. A severe NSD angle (> 15°) showed the strongest independent association with higher-grade pneumatization, particularly in the ACP and PCP, whereas NSD laterality was not independently associated with pneumatization. These findings support the inclusion of NSD angle severity in the radiological assessment of patients undergoing preoperative skull base CT, especially when extensive clinoid pneumatization may influence surgical planning.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13685246/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149227080","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prenatal MRI quantitative analysis of the nonlinear morphological changes of the choroid plexus during the progression of fetal ventriculomegaly.","authors":"Yuhan Zhang, Fengwei Yu, Meng Zeng, Yan Li, Li Zeng, Chen Liu, Ping Cai, Xinwei Li, Wei Chen, Mingshan Du","doi":"10.1186/s12880-026-02601-y","DOIUrl":"10.1186/s12880-026-02601-y","url":null,"abstract":"<p><strong>Background: </strong>To quantify fetal choroid plexus (CP) morphological changes across ventriculomegaly (VM) grades using routine 1.5-T MRI and assess their association with lateral ventricle dilatation.</p><p><strong>Methods: </strong>In this retrospective study, 135 singleton fetuses (24-37 weeks' gestation) were categorized by atrial width (AW): normal, mild VM, moderate VM, severe VM). On coronal T2-weighted images, CP length, width, perimeter, and area were measured. The two-dimensional (2D) CP ratio (CPR-2D) and CP-to-ventricle width ratio were calculated. Analysis of covariance (ANCOVA) adjusted for gestational age and fetal biometrics, and intraclass correlation coefficients (ICC) were used for analysis.</p><p><strong>Results: </strong>Absolute CP measurements exhibited non-linear, stage-specific alterations: CP length and perimeter increased in mild VM, decreased in moderate VM, and increased again in severe VM. CP width progressively decreased until moderate VM before rebounding in severe cases. Conversely, ratio-based indices demonstrated a consistent, monotonic linear decrease as VM severity increased, which remained significant after adjusting for covariates (p < 0.05). While morphological asymmetry existed, the downward trend of ratios remained consistent for both ventricles.</p><p><strong>Conclusions: </strong>The fetal CP undergoes a non-linear change process during the expansion of the lateral ventricle. However, based on ratio-based indicators, especially CPR-2D, stable and reproducible imaging biomarkers can be provided.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13693133/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149242773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AI-assisted ultrasound for trauma: a scoping review.","authors":"Mümin Murat Yazıcı, İsmail Ataş, Enes Hamdioğlu, Cengiz Kazdal, Ali Çelik, Özcan Yavaşi, Özlem Bilir","doi":"10.1186/s12880-026-02603-w","DOIUrl":"10.1186/s12880-026-02603-w","url":null,"abstract":"<p><strong>Objective: </strong>Artificial intelligence (AI) has become a disruptive technology in medicine, with rapidly developing implications for education, diagnosis, image enhancement and assessment, prognosis, and treatment. The literature on AI-assisted ultrasound is still developing. Therefore, we aimed to investigate how AI is employed in ultrasound for trauma and how AI-assisted ultrasound could be deployed in clinical practice.</p><p><strong>Materials and methods: </strong>The review was conducted according to JBI's scoping review methodology. An appropriate search strategy was initially formulated for PubMed, Scopus, Web of Science (WoS), and Google Scholar (backward and forward citation tracking). The concepts of \"artificial intelligence\", \"ultrasound\", and \"trauma\" were searched across these databases using their respective Medical Subject Headings (MeSH). The core concepts were AI applications integrated with ultrasound in trauma. In this context, studies focusing on the following categories relating to the use of AI-assisted ultrasound in the field of trauma have been included in the review: (1) diagnostic accuracy, (2) differential diagnosis and confirmation of diagnosis, (3) quality and standardization, (4) operator support, and (5) training and modelling. Data extraction and analysis were performed using Microsoft Excel and were verified.</p><p><strong>Results: </strong>Twenty-one studies were included. Targeted AI tasks primarily comprised diagnostic abnormality detection and segmentation/localization. Predominantly focused assessment with sonography for trauma (FAST), thoracic trauma evaluation, and extremity trauma. Although high apparent diagnostic accuracy was reported across development-stage retrospective cohorts, independent external validation was executed in only one study, with the remainder bounded by single-center data or synthetic phantoms. Crucially, none of the studies assessed subsequent patient outcomes such as mortality, time to intervention, or resource utilization.</p><p><strong>Conclusion: </strong>This scoping review confirms that while deep learning frameworks show promising retrospective diagnostic performance and image-quality automation, their translation into acute trauma care faces critical technical and clinical bottlenecks. Key current limitations include restricted model interpretability, missing regulatory pathways, a lack of clinician trust, inherent decision-making biases, poor algorithmic transparency, and the lack of patient-outcome tracking. To address these barriers, future studies should be designed with stronger methodological rigor, including prospective, multicenter, randomized controlled, and externally validated approaches.</p><p><strong>Trial registration: </strong>No clinical trial number is available.</p>","PeriodicalId":9020,"journal":{"name":"BMC Medical Imaging","volume":"26 1","pages":""},"PeriodicalIF":3.9,"publicationDate":"2026-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13692931/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149242809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}