Radiology最新文献

{"title":"Invasive Breast Cancers Missed by AI Screening of Mammograms.","authors":"Ok Hee Woo, Sung Eun Song, Su Jin Choe, Minhye Kim, Kyu Ran Cho, Bo Kyoung Seo","doi":"10.1148/radiol.242408","DOIUrl":"10.1148/radiol.242408","url":null,"abstract":"<p><p>Background Little is known about the features of invasive breast cancers missed by artificial intelligence (AI) on mammograms. Purpose To assess the false-negative rate (FNR) of AI mammogram evaluation according to molecular subtype and to investigate the features of and reasons for AI-missed cancers. Materials and Methods This retrospective study identified consecutive patients diagnosed with breast cancer between January 2014 and December 2020. Commercial AI software was used to read the mammograms, and abnormality score (AS) was acquired. AI-missed cancers were defined as those for which AI did not identify a precise location matching the reference standard. The FNR was calculated by counting AI-missed cancers according to molecular subtype (hormone receptor-positive [luminal] vs human epidermal growth factor receptor 2 [HER2]-enriched vs triple-negative). Three blinded radiologists classified AI-missed cancers as either actionable or under threshold, and reasons for misses were determined through nonblinded reviews. Features were compared according to AI detection with the χ² test. Results A total of 1082 consecutive women diagnosed with 1097 cancers (mean age, 54.3 years ± 11 [SD]) were included. AI missed 14% (154 of 1097) of cancers. The FNR was lowest in the HER2-enriched subtype (9% [36 of 398] in the HER2-enriched subtype, 17.2% [106 of 616] in the luminal subtype, and 14.5% [12 of 83] in the triple-negative subtype; <i>P</i> = .001). Compared with AI-detected cancers, AI-missed cancers were associated with younger age, a tumor size less than or equal to 2 cm, a lower histologic grade, fewer lymph node metastases, more Breast Imaging Reporting and Data System category 4 findings, lower Ki-67 expression, and nonmammary zone locations (all, <i>P</i> < .05). In blinded reviews, 61.7% (95 of 154) of AI-missed cancers were actionable; the reasons for misses were dense breasts (<i>n</i> = 56), nonmammary zone locations (<i>n</i> = 22), architectural distortions (<i>n</i> = 12), and amorphous microcalcifications (<i>n</i> = 5). Conclusion To reduce AI-missed cancers on mammograms, attention should be given to luminal cancer, dense breasts, nonmammary zone locations, architectural distortions, and amorphous calcifications. Published under a CC BY 4.0 license. <i>Supplemental material is available for this article.</i> See also the editorial by Mullen in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e242408"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144476523","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":"Bilateral Partial Anomalous Pulmonary Venous Return.","authors":"Limin Lei, Yuhan Zhou","doi":"10.1148/radiol.242606","DOIUrl":"https://doi.org/10.1148/radiol.242606","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e242606"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144258894","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":"Cost-Effectiveness and Efficacy of Noninvasive Colorectal Cancer Screening: An Important Step Toward Widespread Adoption of CT Colonography.","authors":"Samuel J Galgano, Elainea N Smith","doi":"10.1148/radiol.251493","DOIUrl":"https://doi.org/10.1148/radiol.251493","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e251493"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144258896","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":"Congratulations to the 2025 Editorial Fellows.","authors":"Tina Y Poussaint","doi":"10.1148/radiol.259011","DOIUrl":"https://doi.org/10.1148/radiol.259011","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e259011"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144476497","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":"Cumulative Effect of Targeted Interventions on Radiologist Recommendations for Additional Imaging.","authors":"Nooshin Abbasi, Neena Kapoor, Ronilda Lacson, Jeffrey P Guenette, Sonali Desai, David Lucier, Sanjay Saini, Rachel Sisodia, Ali S Raja, David W Bates, Ramin Khorasani","doi":"10.1148/radiol.243750","DOIUrl":"10.1148/radiol.243750","url":null,"abstract":"<p><p>Background Ambiguous or unnecessary radiologist recommendations for additional imaging (RAIs) can lead to excessive imaging use and diagnostic errors. Purpose To determine the cumulative impacts of multifaceted technology-enabled interventions aimed at optimizing RAI on RAI rate, actionability, and resolution over an 8-year period. Materials and Methods In this retrospective cohort study, conducted from January 2015 to December 2022, radiology reports from two tertiary hospitals (study site and control site) were analyzed. A series of quality improvement interventions, including radiologist education, electronic communication tools for tracking RAIs, and performance reports, were implemented at the study site but not at the control site. The RAI rate trend over time was compared between the sites using linear regression. Mixed-effects logistic regression was performed to assess the intervention impact on the RAI rate. RAI actionability and resolution were compared between the sites using the Fisher exact test. <i>P</i> values were corrected using the Bonferroni method. Results Among 7 502 521 total radiology reports (1 323 459 patients) (study site, 3 608 977 reports and 660 051 patients; control site, 3 893 544 reports and 690 115 patients), the RAI rate of the study site decreased by 44%, from 10% (8202 of 81 586) to 5.6% (8972 of 159 599), but remained unchanged at the control site, at 10.9% (8757 of 80 030) vs 11% (16 978 of 153 711) (regression coefficient, -0.09; 95% CI: -0.1, -0.09; <i>P</i> < .001). RAI rates declined with each successive intervention at the study site (<i>P</i> < .001), with regression coefficients decreasing progressively from -0.12 (95% CI: -0.14, -0.10) for the initial intervention to -0.81 (95% CI: -0.83, -0.78) for the final intervention. Recommendation actionability at the study site increased 7.6-fold (from 5.6% [19 of 340] to 42.3% [144 of 340]; <i>P</i> < .001) but remained unchanged at the control site (from 15.0% [51 of 340] to 13.8% [47 of 340]; <i>P</i> = .73). Actionable RAIs were more frequently resolved at the study site than at the control site (84.7% [122 of 144] vs 59.6% [28 of 47]; <i>P</i> < .001). Conclusion Multifaceted interventions to optimize RAI improved the rate, actionability, and resolution of RAI. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Russell in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e243750"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144209337","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":"CT Colonography versus Multitarget Stool DNA Test for Colorectal Cancer Screening: A Cost-Effectiveness Analysis.","authors":"Perry J Pickhardt, Loredana Correale, Cesare Hassan","doi":"10.1148/radiol.243775","DOIUrl":"https://doi.org/10.1148/radiol.243775","url":null,"abstract":"<p><p>Background Colorectal cancer (CRC) is largely preventable or curable with effective screening. Purpose To compare both the clinical efficacy and cost effectiveness of CRC screening with CT colonography (CTC) with those of multitarget stool DNA (mt-sDNA) testing. Materials and Methods A state-transition Markov model was constructed using updated natural history evidence for colorectal polyps applied to a hypothetical 10 000-person cohort representative of the 45-year-old U.S. population. Three screening strategies were modeled with these data: mt-sDNA testing every 3 years, the conventional CTC (CTC_conv) strategy of immediate polypectomy for all polyps measuring at least 6 mm every 5 years, and the surveillance CTC (CTC_surv) strategy of 3-year CTC follow-up for small polyps (6-9 mm) and polypectomy for large polyps (≥10 mm). Multifaceted model validation was performed to confirm robustness. A detailed sensitivity analysis was performed in addition to the base-case scenario. Results Without screening, the cumulative incidence of CRC was 7.5% (<i>n</i> = 752), which was reduced by 59% (<i>n</i> = 310) with mt-sDNA, 75% (<i>n</i> = 190) with CTC_conv, and 70% (<i>n</i> = 223) with CTC_surv screening. The estimated programmatic costs per person for no screening, mt-sDNA, CTC_conv, and CTC_surv were $4955, $6011, $4422, and $3913, respectively. The estimated cost per quality-adjusted life year (QALY) gained for mt-sDNA testing was $8878, whereas both CTC strategies resulted in cost savings. Accordingly, CTC strategies dominated over both mt-sDNA and no screening (ie, more clinically efficacious and more cost-effective [cost-saving]). However, the CTC_conv strategy was not as cost-effective as the CTC_surv strategy, as costs related to more optical colonoscopies did not offset the corresponding small gains in QALYs. The results were similar when CRC screening began at 50 and 65 years of age. Conclusion Both CTC screening strategies dominated over mt-sDNA screening and no screening, yielding cost savings and increased clinical efficacy. The CTC strategy consisting of 3-year surveillance for small colorectal polyps and colonoscopy referral for large polyps achieved the best overall balance of cost and clinical efficacy. © RSNA, 2025 <i>Supplemental material is available for this article.</i> See also the editorial by Galgano and Smith in this issue.</p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e243775"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144258897","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":"Caudate Lobe Paraganglioma of the Liver.","authors":"Binglin Lai, Junyuan Zhong","doi":"10.1148/radiol.242899","DOIUrl":"https://doi.org/10.1148/radiol.242899","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e242899"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144476496","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":"Intradural Disk Herniation.","authors":"Qiang Fang, Jingzhen He","doi":"10.1148/radiol.242553","DOIUrl":"https://doi.org/10.1148/radiol.242553","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e242553"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144317799","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":"A Data-Centric Approach to Deep Learning for Brain Metastasis Analysis at MRI.","authors":"Laurens Topff, Liliana Petrychenko, Neeraj Jain, Sara Lingier, Jeroen Bertels, Patricio Astudillo, Milan Prosec, Pablo Menéndez Fernández-Miranda, Olivier Gevaert, Marion Smits, Sophie Derks, Eline Verhaak, Patrick E J Hanssens, Enrique Marco de Lucas, Rodrigo Sutil, Pablo D Dominguez, Adina Negoita, Ernst Visser, David Corral Fontecha, Loes M M Braun, Dieta Brandsma, Jacob J Visser, Erik R Ranschaert, Kevin B W Groot Lipman, Regina G H Beets-Tan","doi":"10.1148/radiol.242416","DOIUrl":"10.1148/radiol.242416","url":null,"abstract":"<p><p>Background With the increasing incidence of brain metastases (BMs), artificial intelligence models have shown promise in assisting with the detection and volumetric analysis of lesions at MRI. However, current models are limited in identifying small lesions and lack generalizability. Purpose To develop a generalizable deep learning system for detecting, segmenting, and longitudinally tracking BMs of any size at MRI. Materials and Methods In this retrospective study, a data-centric approach to deep learning model development was used. A multicenter dataset was collected, comprising pre- and/or posttreatment MRI scans from patients with BMs and MRI scans from patients with cancer without BMs (December 2015 to August 2023). Iterative data annotation by radiologists with systematic quality control increased the consistency of reference segmentations. A modified nnU-Net framework, with robust data preprocessing and augmentation, was used. Lesion-wise detection metrics and segmentation performance, Dice similarity coefficient, and normalized surface distance were evaluated. Results In total, 1985 scans from 1623 patients (mean age, 62.0 years ± 12.2 [SD]; 743 female patients, 157 patients of unknown sex), with 5552 BMs, were included. BMs were present in 64.8% of the scans (1286 of 1985), 36.0% (463 of 1286) of which were posttreatment scans. The model was trained on 1451 scans acquired on 30 different scanners. In internal testing (<i>n</i> = 223), sensitivity was 98.0% (95% CI: 96.3, 99.0; 449 of 458 lesions). In external testing (<i>n</i> = 311), sensitivity was 97.4% (95% CI: 96.2, 98.2; 935 of 960; <i>P</i> = .58), with a mean of 0.6 false positives per patient. The sensitivity remained high for all lesion sizes, including those less than 3 mm in diameter (93.3% [95% CI: 89.1, 96.0]; 196 of 210). Median Dice similarity coefficient was 0.89 and 0.90 for the internal and external test datasets, respectively (<i>P</i> = .13). Median normalized surface distance was 0.99 for both datasets. Conclusion The deep learning system demonstrated high performance and generalizability in detecting and segmenting BMs of all sizes on pre- and posttreatment MRI scans. © RSNA, 2025 <i>Supplemental material is available for this article.</i></p>","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e242416"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12207652/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144476493","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Closing the Loop: Improving Patient Care When Imaging Follow-up is Needed.","authors":"Eric J Russell","doi":"10.1148/radiol.251369","DOIUrl":"https://doi.org/10.1148/radiol.251369","url":null,"abstract":"","PeriodicalId":20896,"journal":{"name":"Radiology","volume":"315 3","pages":"e251369"},"PeriodicalIF":12.1,"publicationDate":"2025-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144209336","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}