Radiology最新文献

{"title":"Beyond Spirometry: AI-Driven Chest CT Analysis, a New Frontier in Monitoring Amyotrophic Lateral Sclerosis.","authors":"Amir Ali Rahsepar, Arash Bedayat","doi":"10.1148/radiol.250988","DOIUrl":"https://doi.org/10.1148/radiol.250988","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 2","pages":"e250988"},"PeriodicalIF":12.1,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144034639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multireader Diagnostic Accuracy of Abbreviated Breast MRI for Screening Women with Extremely Dense Breasts.","authors":"Sophie E L van Grinsven, Ritse M Mann, Evelyn M Monninkhof, Katya Duvivier, Mathijn D F de Jong, Petra K de Koekkoek-Doll, Claudette E Loo, Ruud M Pijnappel, Rogier van der Sluijs, Jeroen Veltman, Carla H van Gils, Wouter B Veldhuis","doi":"10.1148/radiol.241233","DOIUrl":"https://doi.org/10.1148/radiol.241233","url":null,"abstract":"<p><p>Background Abbreviated MRI may reduce costs and time of supplemental breast cancer screening. The Dense Tissue and Early Breast Neoplasm Screening (DENSE) trial provides an opportunity to study this protocol in a true screening population. Purpose To compare multireader diagnostic accuracy of various abbreviated screening breast MRI protocols with that of the full multiparametric protocol in the DENSE trial, focusing on identifying the minimal protocol necessary to maintain high diagnostic accuracy. Materials and Methods In this secondary analysis of a subset from the DENSE trial, seven radiologists performed incremental readings of first-round screening MRI scans in women with extremely dense breasts and findings negative for cancer at mammography. Different sequences were added in four incremental steps. The first step included both high-temporal low-spatial and low-temporal high-spatial dynamic T1-weighted series, up to 120 seconds after contrast agent injection. The final step added all full-protocol sequences. Each radiologist assessed the same MRI scans and provided Breast Imaging Reporting and Data System scores for all four incremental steps. Pooled sensitivity and specificity were calculated across all readers per step using a generalized estimating equation model, and pooled reading time per step was calculated using a linear mixed model. Results The first-round screening included 518 MRI scans from 518 women (median age, 53 years; IQR, 51-59 years), including 83 breast cancers: 68 invasive cancers (82%) and 15 ductal carcinomas in situ (18%). There was no evidence of a difference in sensitivity between abbreviated protocol (84.3%; 95% CI: 77.7, 89.2) and the full multiparametric MRI protocol (85.9%; 95% CI: 80.0, 90.3; <i>P</i> = .68). There was also no evidence of a difference in specificity between abbreviated protocol (73.9%; 95% CI: 70.7, 76.9) and full protocol (75.8%; 95% CI: 72.8, 78.5; <i>P</i> = .39). The abbreviated protocol had a pooled reading time (49.7 seconds; 95% CI: 48.5, 50.9) that was almost 50% shorter than the full protocol (96.4 seconds; 95% CI: 94.3, 98.5; <i>P</i> < .001) with 70%-80% shorter scanning time, depending on hospital and scanner vendor. Conclusion In women in the DENSE trial with extremely dense breasts and findings that were negative for cancer at mammography, abbreviated breast MRI for first-round screening had high diagnostic accuracy that was comparable to full multiparametric protocol, at much shorter reading and scanning times. ClinicalTrials.gov Identifier: NCT01315015 © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Kataoka and Honda in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 2","pages":"e241233"},"PeriodicalIF":12.1,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144209333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Improved Filtering for CT Perfusion in Stroke: From a Small Step to a Big Leap.","authors":"Alexandre Krainik","doi":"10.1148/radiol.250764","DOIUrl":"https://doi.org/10.1148/radiol.250764","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 2","pages":"e250764"},"PeriodicalIF":12.1,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143981126","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimizing an Abbreviated MRI Protocol for Women with Dense Breasts.","authors":"Masako Kataoka, Maya Honda","doi":"10.1148/radiol.250981","DOIUrl":"https://doi.org/10.1148/radiol.250981","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 2","pages":"e250981"},"PeriodicalIF":12.1,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144111800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical Outcomes of Interstitial Lung Abnormalities Detected in the Korean National Lung Cancer CT Screening Program.","authors":"Hyungjin Kim, Eunseo Jo, Jinseob Kim, Soon Ho Yoon, Seung Ho Choi, Yoosoo Chang, Seungho Ryu, Jin Mo Goo","doi":"10.1148/radiol.243651","DOIUrl":"https://doi.org/10.1148/radiol.243651","url":null,"abstract":"<p><p>Background Limited evidence exists on the prevalence and outcomes of interstitial lung abnormalities (ILAs) in lung cancer screening populations, particularly Asian populations. Purpose To investigate the prevalence of ILAs and the association of ILAs with lung cancer, idiopathic pulmonary fibrosis (IPF), and mortality outcomes in an Asian population. Materials and Methods In this nationwide, population-based retrospective study, baseline screenings from the Korean National Lung Cancer Screening Program performed between August 2019 and December 2020 were analyzed. ILAs were identified from CT structured reports based on program radiologists' visual assessment, and ILA prevalence was analyzed across age groups. Incidence rate ratios were calculated for lung cancer incidence, IPF, and all-cause mortality comparing individuals with ILAs versus individuals without ILAs, and multivariable Cox regression analyses were performed to examine associations between ILAs and these outcomes. Results Among 125 600 individuals (mean age, 62 years ± 5.3 [SD]; 123 331 men), ILA prevalence was 2.65% (3324 of 125 600) and was strongly associated with older age (<i>P</i> < .001). The lung cancer incidence rate was higher in the ILA group (2009 vs 412 per 100 000 person-years, <i>P</i> < .001; incidence rate ratio, 4.88), as was the all-cause mortality rate (2334 vs 712 per 100 000 person-years, <i>P</i> < .001; incidence rate ratio, 3.28). During a median follow-up of 2.9 years, IPF was diagnosed in 3.55% (118 of 3324) of individuals with ILAs (incidence rate, 1344 per 100 000 person-years in group with ILAs vs 18 per 100 000 person-years in group without ILAs, <i>P</i> < .001; incidence rate ratio, 73.24). In multivariable analyses, individuals with ILAs had a threefold higher risk of lung cancer (adjusted hazard ratio, 3.18 [95% CI: 2.71, 3.73]; <i>P</i> < .001) and twofold higher all-cause mortality (adjusted hazard ratio, 2.37 [95% CI: 2.09, 2.69]; <i>P</i> < .001). Individuals with ILAs showed a markedly higher risk of IPF diagnosis, with more than 60-fold higher risk (adjusted hazard ratio, 63.4 [95% CI: 45.9, 87.7]; <i>P</i> < .001). Conclusion The presence of ILAs was associated with higher risks of lung cancer, IPF, and mortality. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Baruah and Kabakus in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 2","pages":"e243651"},"PeriodicalIF":12.1,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144151445","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of Background Parenchymal Enhancement on Diagnostic Performance of Breast MRI: A Systematic Review and Meta-Analysis.","authors":"Sonja Bechyna, Pascal A T Baltzer","doi":"10.1148/radiol.241919","DOIUrl":"10.1148/radiol.241919","url":null,"abstract":"<p><p>Background Breast MRI is the most sensitive modality for detecting breast cancer, but its diagnostic performance can be reduced by background parenchymal enhancement (BPE). However, the implications of the degree of BPE in diagnostic evaluation remain controversial due to conflicting evidence. Purpose To conduct a systematic review and meta-analysis to determine whether the presence of moderate or marked BPE in women undergoing breast MRI negatively affects diagnostic performance compared with minimal or mild BPE. Materials and Methods On May 6, 2024, a systematic review of PubMed and Google Scholar was conducted on studies assessing the diagnostic performance of contrast-enhanced MRI in detecting breast cancer. Studies published until May 2024 were included if they provided raw data to extract or calculate true-positive, false-positive, true-negative, and false-negative results. Data were integrated using a bivariate random-effects model, and the sensitivity and specificity were compared between minimal or mild and moderate or marked BPE. Sources of heterogeneity were explored using meta-regression analysis. Summary receiver operating characteristic analysis was performed to compare overall diagnostic performance at different BPE levels by using a <i>Z</i> test. Results Eight studies of 6044 women (mean age, 52 years) were included. Minimal or mild BPE was associated with higher sensitivity (95.3% vs 84.1%; <i>P</i> = .02) and higher specificity (89.0% vs 78.7%; <i>P</i> = .02) compared with moderate or marked BPE. Diagnostic performance was higher for minimal or mild BPE (area under the receiver operating characteristic curve [AUC], 0.95; 95% CI: 0.92, 0.96) compared with moderate or marked BPE (AUC, 0.91; 95% CI: 0.88,0.93; <i>Z</i> score, 2.69; <i>P</i> = .007). Moderate or marked BPE and publication year in cases of moderate or marked BPE were covariates influencing the diagnostic odds ratio. Moderate or marked BPE remained an independent predictor of the diagnostic performance at multivariable analysis (diagnostic odds ratio, -1.33; <i>P</i> = .002). Conclusion The degree of BPE influences the diagnostic performance of breast MRI, with minimal or mild BPE being associated with higher sensitivity and specificity and moderate or marked BPE being associated with lower diagnostic performance. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Bahl in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 2","pages":"e241919"},"PeriodicalIF":12.1,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144151488","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessment of the Rotator Cuff Muscles: State-of-the-Art MRI and Clinical Implications.","authors":"Georg C Feuerriegel,Benjamin Fritz,Adrian A Marth,Stefan Sommer,Karl Wieser,Reto Sutter","doi":"10.1148/radiol.242131","DOIUrl":"https://doi.org/10.1148/radiol.242131","url":null,"abstract":"Rotator cuff (RC) tears are a common cause of shoulder pain and, depending on the tear pattern, can substantially limit daily activities and affect quality of life. After a tendon tear, the RC muscle undergoes degenerative changes, including fatty degeneration and volume atrophy. The extent of fatty muscle degeneration has been associated with poorer functional and surgical outcomes after RC reconstruction, and the evaluation of the RC muscles is one of the most important factors for deciding whether to perform an RC repair. A variety of qualitative and quantitative methods are available to assess RC fatty muscle degeneration and volume atrophy based on MRI. This review provides a detailed overview of the RC muscle assessment, including qualitative measurements to assess the RC muscles, such as Goutallier grading, tangent sign, and fish backbone sign. Another focus is state-of-the-art quantitative MRI techniques, including chemical shift-based techniques such as single-voxel MR spectroscopy, multipoint water-fat separation, and MR relaxometry. Furthermore, the clinical relevance of qualitative and quantitative intramuscular fat assessment is addressed, and its potential importance for treatment planning and patient outcomes is discussed.","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"13 1","pages":"e242131"},"PeriodicalIF":19.7,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143914843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Cinematic Rendering of Superior Mesenteric Arteriovenous Fistula.","authors":"Wendi Kang,Jijun Zhang","doi":"10.1148/radiol.242237","DOIUrl":"https://doi.org/10.1148/radiol.242237","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"53 1","pages":"e242237"},"PeriodicalIF":19.7,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143914851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Anatomy-derived 3D Aortic Hemodynamics Using Fluid Physics-informed Deep Learning.","authors":"Haben Berhane,Anthony Maroun,David Dushfunian,Justin Baraboo,Gabriela Martinez,Tyler Jacobson,Ulas Bagci,Bradley D Allen,Michael Markl","doi":"10.1148/radiol.240714","DOIUrl":"https://doi.org/10.1148/radiol.240714","url":null,"abstract":"Background Four-dimensional (4D) flow MRI provides assessment of thoracic aorta hemodynamic measures that are increasingly recognized as important biomarkers for risk assessment. However, long acquisition times and cumbersome data analysis limit widespread availability. Purpose To evaluate the feasibility and accuracy of a generative artificial intelligence (AI) approach (fluid physics-informed cycle generative adversarial network [FPI-CycleGAN]) in quantifying aorta hemodynamics directly from anatomic input as an alternative to 4D flow MRI. Materials and Methods Patients were retrospectively identified from a dataset of clinical cardiothoracic MRI examinations performed between November 2011 and July 2020. All patients underwent aortic 4D flow MRI, which served as a reference standard for training and testing of FPI-CycleGANs. A three-dimensional (3D) segmentation of the aortic geometry was used as the only input to predict systolic aortic hemodynamics, with separate networks for bicuspid aortic valve (BAV) (994 in the training set and 248 in the test set) and tricuspid aortic valve (TAV) (419 in the training set and 104 in the test set). Voxel-by-voxel and regional analyses were used to quantify and compare (AI vs the reference standard, 4D flow) systolic velocity vector fields, peak velocity, wall shear stress (WSS), and classification of aortic valve stenosis. Results In total, 1765 patients (median age, 53 years [IQR, 41-63 years]; 1242 patients had BAV and 523 had TAV) were included. Mean AI computation time was 0.15 second ± 0.11 (SD), and total training was 1500 and 3600 minutes for the TAV and BAV networks, respectively. The FPI-CycleGAN predicted systolic 3D velocity vector fields accurately, with low bias (<0.01 m/sec) and excellent limits of agreements (±0.06-0.08 m/sec). For peak velocities and WSS, there was strong agreement between FPI-CycleGAN and 4D flow (r2 = 0.930-0.957 [P < .001], with relative differences of 6.2%-9.8%). AI accurately classified aortic valve stenosis severity in 85.8% of patients (302 of 352) (κ = 0.80 [95% CI: 0.71, 0.89]). The FPI-CycleGAN was robust to one- and two-voxel dilation and erosion (bias, -0.05 to 0.1 m/sec) and ±5° rotation (bias, -0.02 to 0.03 m/sec) of the input data. The application of the trained FPI-CycleGAN in an external test set with contrast-enhanced MR angiography (n = 60 patients) as AI input data demonstrated strong to excellent performance for peak velocities and WSS (r2 = 0.944-0.965 [P < .001], with relative differences of 6.2%-9.2%). Conclusion Aorta 3D hemodynamics can be derived from anatomic input in less than 1 second using an FPI-CycleGAN and demonstrate strong agreement with in vivo 4D flow MRI systolic hemodynamics. © RSNA, 2025 Supplemental material is available for this article.","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"115 1","pages":"e240714"},"PeriodicalIF":19.7,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143914861","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Is AI for CAD-RADS 2.0 Ready for Prime Time?","authors":"Marly van Assen, Carlo N De Cecco","doi":"10.1148/radiol.251210","DOIUrl":"https://doi.org/10.1148/radiol.251210","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 2","pages":"e251210"},"PeriodicalIF":12.1,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144005870","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}