Radiology-Artificial Intelligence最新文献

筛选
英文 中文
Evaluation of a Rapid and Practical Deep Learning Framework for T1-weighted MRI-based Brain Age Estimation. 基于t1加权mri的脑年龄估计快速实用深度学习框架的评估。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-11-11 DOI: 10.1148/ryai.260310
Hyun Woo Kim, Hyun Woong Roh, Chang Hyung Hong, Sang Joon Son
{"title":"Evaluation of a Rapid and Practical Deep Learning Framework for T1-weighted MRI-based Brain Age Estimation.","authors":"Hyun Woo Kim, Hyun Woong Roh, Chang Hyung Hong, Sang Joon Son","doi":"10.1148/ryai.260310","DOIUrl":"https://doi.org/10.1148/ryai.260310","url":null,"abstract":"<p><p>Purpose To develop and evaluate a rapid brain-age estimation framework using freely available Python tools and assess the clinical relevance of brain-age gap (BAG). Materials and Methods In this retrospective study conducted from March 2025 to August 2026, T1-weighted MR images from cohorts of cognitively normal adults (<i>n</i> = 4,346) were divided into training, validation, and internal test sets; two cohorts (<i>n</i> = 315 and 209) served as external test sets. Clinical evaluation used the BICWALZS cohort (<i>n</i> = 954; cognitively normal and various dementias). Preprocessing used freely available Python tools only. The model was trained to predict chronological age as the reference standard. Associations of the residualized BAG with diagnosis, amyloid PET positivity, white matter hyperintensity (WMH) severity, stroke history, and longitudinal clinical worsening were evaluated using ANCOVA, logistic regression, and Cox models. Results The development cohort included 4,346 cognitively normal adults (mean age, 47.0 ± 21.2 years; 2,414 women). Chronological age prediction yielded R<sup>2</sup> = 0.95 and MAE, 3.76 years in the internal test set and R<sup>2</sup> = 0.69-0.90 and MAE, 4.82-5.15 years in external test sets. The residualized BAG increased across diagnostic groups (<i>P</i> < .001) and was associated with WMH severity (odds ratio [OR], 1.64; <i>P</i> < .001) and stroke history (OR, 1.46; <i>P</i> < .001), but not with amyloid positivity after full adjustment (<i>P</i> = .17), and predicted longitudinal clinical worsening (HR, 1.92). Conclusion The framework accurately predicted age, while residualized BAG was associated with diagnosis, cerebrovascular burden, and longitudinal clinical worsening. © RSNA, 2026.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260310"},"PeriodicalIF":20.1,"publicationDate":"2026-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149388002","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Anatomy-specific Performance of CTA-based AI for Anterior Circulation Occlusion: Systematic Review and Meta-analysis. 基于CT血管造影的人工智能在前循环闭塞中的解剖特异性表现:系统回顾和荟萃分析。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-11-01 DOI: 10.1148/ryai.260261
Yike Liu, Chaofan Li, Zepeng Ren, Yuchen Hu, Chaohua Wang, Chunyan Chen, Xijin Liu
{"title":"Anatomy-specific Performance of CTA-based AI for Anterior Circulation Occlusion: Systematic Review and Meta-analysis.","authors":"Yike Liu, Chaofan Li, Zepeng Ren, Yuchen Hu, Chaohua Wang, Chunyan Chen, Xijin Liu","doi":"10.1148/ryai.260261","DOIUrl":"10.1148/ryai.260261","url":null,"abstract":"<p><p>Purpose To evaluate anatomy-specific diagnostic performance of CT angiography (CTA)-based artificial intelligence (AI) for anterior circulation occlusion detection and negative-result implications for distal occlusions. Materials and Methods In this Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic Test Accuracy Studies-compliant, prospectively registered systematic review and diagnostic meta-analysis, PubMed, Embase, and Web of Science were searched for AI-based CTA studies from January 1, 2019, through March 11, 2026. Pooled sensitivity and specificity were estimated for global large-vessel occlusions (LVOs), combined internal carotid artery and first segment of the middle cerebral artery (ICA/M1) occlusions, and distal second and third segments of the middle cerebral artery (M2/M3) occlusions with a bivariate random-effects model. Evidence certainty was assessed with Grading of Recommendations Assessment, Development and Evaluation. A supportive, record-level, multilevel bivariate generalized linear mixed model adjusted for occlusion territory, study design, algorithm type, publication year, and section thickness. Results Thirty-one reports involving 15 708 patients were included. Pooled sensitivity and specificity were 80.1% and 91.9% for global LVOs, 91.6% and 92.7% for ICA/M1 occlusions, and 52.7% and 94.8% for M2/M3 occlusions. In the supportive adjusted model, ICA/M1 occlusions had higher sensitivity than global LVOs (odds ratio, 2.28; <i>P</i> < .001), whereas M2/M3 occlusions had lower sensitivity (odds ratio, 0.25; <i>P</i> < .001). Evidence certainty was moderate for ICA/M1 sensitivity and very low for M2/M3 sensitivity. For M2/M3 occlusions, a negative AI result yielded a negative likelihood ratio of 0.50 (95% CI: 0.28, 0.73), corresponding to posttest probabilities of 17.6% and 33.3% at 30% and 50% pretest probabilities. Conclusion CTA-based AI performance varied by occlusion territory; lower M2/M3 sensitivity limited negative-result reliability. <b>Keywords:</b> Computer Aided Diagnosis (CAD), CT-Angiography, Neural Networks, Vascular, Acute Ischemic Stroke, Artificial Intelligence, Anterior Circulation Occlusion, Diagnostic Accuracy, Workflow Triage, Meta-Analysis <i>Supplemental material is available for this article.</i> © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license. See also the commentary by Suri in this issue.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260261"},"PeriodicalIF":20.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148621574","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
False Reassurances from False Negatives: AI Stroke Algorithms and Their Anatomical Blind Spots. 假阴性的假保证:人工智能笔画算法及其解剖盲点。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-11-01 DOI: 10.1148/ryai.260969
Abhinav Suri
{"title":"False Reassurances from False Negatives: AI Stroke Algorithms and Their Anatomical Blind Spots.","authors":"Abhinav Suri","doi":"10.1148/ryai.260969","DOIUrl":"10.1148/ryai.260969","url":null,"abstract":"","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":"8 6","pages":"e260969"},"PeriodicalIF":20.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149325240","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
ARNAI: Artifact Removal Network Based on Autoencoding and Inpainting for Robust Spinal Image Segmentation and Measurement. 基于自编码和图像修复的伪影去除网络在脊柱图像分割和测量中的应用。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-11-01 DOI: 10.1148/ryai.251134
Sang-Jin Park, Jinyoung Choi, Seokwon Kim, Seungeon Song, Insu Park, Dougho Park, Taeyeon Kim, Youjin Lee, Donghoon Yang, Jae Man Cho, Joong Won Yang, Mansu Kim, Heum Dai Kwon, Hong-Gyu Baek, Dae-Chul Cho, Injung Kim
{"title":"ARNAI: Artifact Removal Network Based on Autoencoding and Inpainting for Robust Spinal Image Segmentation and Measurement.","authors":"Sang-Jin Park, Jinyoung Choi, Seokwon Kim, Seungeon Song, Insu Park, Dougho Park, Taeyeon Kim, Youjin Lee, Donghoon Yang, Jae Man Cho, Joong Won Yang, Mansu Kim, Heum Dai Kwon, Hong-Gyu Baek, Dae-Chul Cho, Injung Kim","doi":"10.1148/ryai.251134","DOIUrl":"10.1148/ryai.251134","url":null,"abstract":"<p><p>Purpose To develop an artificial intelligence framework, robust to spinal implants, for automated measurement of spinopelvic parameters on postoperative radiographs. Materials and Methods Lateral lumbar spine radiographs from two institutions (internal dataset [<i>n</i> = 2486]: January 2017-December 2024; external dataset [<i>n</i> = 217]: October 2021-September 2025) were retrospectively reviewed. The Restore, Segment, and Measure (RSM) framework was developed, incorporating a novel Artifact Removal Network based on Autoencoding and Inpainting (ARNAI) to mitigate implant-related artifacts. Segmentation and spinopelvic parameter (including segmental Cobb angle [SCA]) measurement performance relative to expert reference measurements were assessed using the Dice similarity coefficient (DSC), mean absolute error (MAE), Wilcoxon signed rank tests, and Benjamini-Hochberg correction for multiple comparisons. Results With ARNAI added to the segmentation pipeline, the mean DSC increased to 0.870 from 0.814, with marked gains at L3-L5. For implant-containing radiographs in the internal test set, rejected radiographs decreased by 64.21% (95 to 34) and the L4-L5 SCA MAE decreased to 4.7° from 16.2°-15.6° (approximately 70% reduction); this difference remained significant after correction. For the external test set, rejected radiographs decreased by 65.45% (110 to 38) and the L4-L5 SCA MAE decreased to 9.5°-9.7° from 14.2°-14.5° (approximately 33% reduction); however, this improvement did not remain significant after correction, and agreement with expert references remained limited. Conclusion The RSM framework improved automated spinopelvic parameter measurement in implant-containing postoperative radiographs, with the greatest benefit observed for L4-L5 SCA estimation in the internal dataset. In the external dataset, error was reduced but agreement remained limited. <b>Keywords:</b> Convolutional Neural Network, Reconstruction Algorithms, Neural Networks, Segmentation <i>Supplemental material is available for this article.</i> ©RSNA, 2026.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e251134"},"PeriodicalIF":20.1,"publicationDate":"2026-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149128581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Anatomy-informed CT Spatial Phenotype Associated with Extrapancreatic Perineural Invasion in Resected Pancreatic Ductal Adenocarcinoma. 解剖信息的CT空间表型与胰腺导管腺癌切除后胰腺外神经周围浸润相关。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-10-28 DOI: 10.1148/ryai.260357
Chengwei Chen, Jieyu Yu, Yixuan Shen, Danqun Zheng, Yilun Zheng, Fang Liu, Tiegong Wang, Xu Fang, Jing Li, Jianping Lu, Hui Jiang, Chengwei Shao, Yun Bian
{"title":"Anatomy-informed CT Spatial Phenotype Associated with Extrapancreatic Perineural Invasion in Resected Pancreatic Ductal Adenocarcinoma.","authors":"Chengwei Chen, Jieyu Yu, Yixuan Shen, Danqun Zheng, Yilun Zheng, Fang Liu, Tiegong Wang, Xu Fang, Jing Li, Jianping Lu, Hui Jiang, Chengwei Shao, Yun Bian","doi":"10.1148/ryai.260357","DOIUrl":"https://doi.org/10.1148/ryai.260357","url":null,"abstract":"<p><p>Purpose To develop and evaluate an interpretable CT spatial phenotype associated with extrapancreatic perineural invasion (EPNI) in resected pancreatic ductal adenocarcinoma. Materials and Methods This multicenter study included training, internal testing, neoadjuvant testing, retrospective external testing, prospective external testing, and The Cancer Imaging Archive (TCIA) sets. The Preoperative Risk of Perineural High-risk Extension in Tumors (PROPHET) model and threshold were locked before testing. Performance was assessed with area under the receiver operating characteristic curve (AUC) and threshold metrics. Six radiologists compared unaided and assisted readings. A stratified Cox model related PROPHET risk to survival, adjusting for age, sex, tumor size, carbohydrate antigen 19-9, nodal category, and differentiation. Results Among 1854 patients (median age, 63 years; 1131 male), AUCs were 0.92 (95% CI, 0.86-0.96) in prospective external testing, 0.85 (95% CI, 0.75-0.93) in retrospective external testing, 0.90 (95% CI, 0.83-0.95) after neoadjuvant therapy, 0.99 (95% CI, 0.98-> 0.99) in internal testing, and 0.80 (95% CI, 0.68-0.93) in TCIA. TCIA sensitivity and specificity were 24/30 (80%) and 2/19 (11%). Assistance increased mean sensitivity to 93% from 82% and reduced reading time to 3.02 from 3.64 minutes per patient (both Holm-adjusted <i>P</i> < .001). Higher PROPHET risk was associated with shorter survival (adjusted hazard ratio, 1.95; 95% CI, 1.53-2.47; nominal <i>P</i> < .001). Conclusion PROPHET provides a CT-visible tumor-vessel spatial phenotype associated with pathologically confirmed EPNI. © The Author(s) 2026. Published by the Radiological Society of North America under a CC BY 4.0 license.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260357"},"PeriodicalIF":20.1,"publicationDate":"2026-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149288931","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Automation Bias and Deskilling in Radiology AI: A Confidence-stratified Triage Model for Workflow Integration. 放射学人工智能中的自动化偏差和去技能化:工作流集成的信心分层分类模型。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-10-28 DOI: 10.1148/ryai.260591
Nalan Karunanayake, Javin Schefflein, Binsheng Zhao, Lawrence H Schwartz
{"title":"Automation Bias and Deskilling in Radiology AI: A Confidence-stratified Triage Model for Workflow Integration.","authors":"Nalan Karunanayake, Javin Schefflein, Binsheng Zhao, Lawrence H Schwartz","doi":"10.1148/ryai.260591","DOIUrl":"https://doi.org/10.1148/ryai.260591","url":null,"abstract":"<p><p>Radiology is adopting artificial intelligence (AI) faster than it is studying what AI does to radiologists' perceptual expertise and interpretive skill. The question has partly shifted from whether these tools perform well on benchmark tasks to what happens to clinical judgment when use of these tools becomes routine. Two risks accompany this transition: automation bias, the tendency to follow AI suggestions without adequate scrutiny, and deskilling, the erosion of perceptual and interpretive ability through disuse, typically revealed when AI support is withdrawn. Incorrect AI output degrades accuracy regardless of experience, the benefit of assistance varies unpredictably across readers and tasks, and performance falls after AI withdrawal; neither risk has been studied longitudinally in radiology. A confidence-stratified triage model is proposed in which calibrated AI confidence, not findings alone, routes cases to low-, intermediate-, or high-confidence streams: output withheld until independent reading, shown with uncertainty signals and a checklist, or shown with random blinded audit. Demonstrated local confidence calibration is a precondition for deployment, and stream thresholds and audit rates should be weighted by task and potential clinical harm, not error probability alone. Rather than applying one workflow to every case, the model redistributes radiologist attention to where AI error is most likely and most consequential, operationalizing four principles: keeping the radiologist in the decision loop, making AI uncertainty visible, inserting targeted cognitive friction, and governing the human-AI interaction as a system. A research agenda to determine whether these countermeasures reduce automation bias and skill erosion in practice is also proposed.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260591"},"PeriodicalIF":20.1,"publicationDate":"2026-10-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149289103","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
TRAM-AI: A Framework for Transparency in Reporting of Medical Imaging Datasets for Artificial Intelligence Development. TRAM-AI:用于人工智能发展的医学成像数据集报告透明度框架。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-10-21 DOI: 10.1148/ryai.260538
Kartik Bose, Pankaj Gupta, Lisa Adams, Rupesh Agrawal, Mariam Aboian, Soumen Basu, Søren Brunak, Ken Chang, Linda C Chu, Renato Cuocolo, Tugba Akinci D'Antonoli, Salvatore Claudio Fanni, Eduardo Moreno Judice de Mattos Farina, Anthony A Gatti, Satyam Ghodasara, Ricardo A Gonzales, Amit Gupta, Katharina Hoebel, William Hsu, Zekun Jiang, Bardia Khosravi, Michail E Klontzas, Agklinta Kiosia, Neeraj Lalwani, Curtis Langlotz, Hoyeon Lee, Linda Moy, Debnath Pal, Iwan Paolucci, Cody Heath Savage, Krithika Rangarajan, Ronnie Sebro, Abhinav Suri, Ali Tejani, Shuncong Wang, Charles E Kahn
{"title":"TRAM-AI: A Framework for Transparency in Reporting of Medical Imaging Datasets for Artificial Intelligence Development.","authors":"Kartik Bose, Pankaj Gupta, Lisa Adams, Rupesh Agrawal, Mariam Aboian, Soumen Basu, Søren Brunak, Ken Chang, Linda C Chu, Renato Cuocolo, Tugba Akinci D'Antonoli, Salvatore Claudio Fanni, Eduardo Moreno Judice de Mattos Farina, Anthony A Gatti, Satyam Ghodasara, Ricardo A Gonzales, Amit Gupta, Katharina Hoebel, William Hsu, Zekun Jiang, Bardia Khosravi, Michail E Klontzas, Agklinta Kiosia, Neeraj Lalwani, Curtis Langlotz, Hoyeon Lee, Linda Moy, Debnath Pal, Iwan Paolucci, Cody Heath Savage, Krithika Rangarajan, Ronnie Sebro, Abhinav Suri, Ali Tejani, Shuncong Wang, Charles E Kahn","doi":"10.1148/ryai.260538","DOIUrl":"https://doi.org/10.1148/ryai.260538","url":null,"abstract":"<p><p>Purpose To develop and evaluate TRAM-AI, a consensus-based checklist for verifying whether medical imaging datasets described as publicly accessible are functionally usable for AI development. Materials and Methods A modified Delphi study was conducted. A steering committee generated candidate checklist items through a targeted literature review. An international expert panel (<i>n</i> = 33) rated items on criticality and feasibility across two Delphi rounds. Consensus was defined as ≥ 75% agreement on both dimensions. The checklist was pilot-validated in 10 publicly available datasets. An advisory group of four experts from Latin America and sub-Saharan Africa evaluated cross-regional applicability. Results Of 16 candidate items, 14 were retained in the final checklist; one item was removed because of feasibility concerns, and another was merged with an existing item because of conceptual redundancy. All 14 final items achieved consensus for both criticality and feasibility. The final TRAM-AI checklist comprises 14 items across four domains: Dataset Access and Persistence (3 items), Technical Interoperability (4 items), Metadata and Reproducibility (4 items), and Ethical, Legal, and Maintenance considerations (3 items). In pilot validation, all items were applicable to all 10 datasets, compliance was 90% for trusted repository, de-dentification, and subject-level mapping, 10% for archive validation, and 30% for partition keys. All four advisory group members endorsed the checklist. Conclusion TRAM-AI provides a practical framework for verifying the functional accessibility of medical imaging datasets, complementing existing methodological reporting standards. Further evaluation and broader validation across diverse geographic settings are warranted. ©RSNA, 2026.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260538"},"PeriodicalIF":20.1,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149239622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Interpretable Deep Learning for Opportunistic Screening of Left Ventricular Dysfunction Screening Using Chest Radiographs. 可解释的深度学习在利用胸片进行左心室功能障碍筛查中的应用。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-10-21 DOI: 10.1148/ryai.260161
Sowon Jang, Dong Hyun Lee, Hye Soo Cho, Sun-Ju Byeon, Bo Yun Hur, Chang Min Park, Minyoung Chung, Ju Gang Nam
{"title":"Interpretable Deep Learning for Opportunistic Screening of Left Ventricular Dysfunction Screening Using Chest Radiographs.","authors":"Sowon Jang, Dong Hyun Lee, Hye Soo Cho, Sun-Ju Byeon, Bo Yun Hur, Chang Min Park, Minyoung Chung, Ju Gang Nam","doi":"10.1148/ryai.260161","DOIUrl":"https://doi.org/10.1148/ryai.260161","url":null,"abstract":"<p><p>Purpose To develop and evaluate a deep learning model (DLLV) for identifying reduced left ventricular ejection fraction (LVEF) on chest radiographs (CXRs) and assess its potential as a prescreening tool for selective transthoracic echocardiography (TTE) in health checkup populations. Materials and Methods Same-day CXR-TTE pairs obtained between January 2016 and September 2021 were retrospectively collected for model development. Counterfactual learning was applied to enhance model performance and interpretability. External testing included one tertiary referral center cohort and two health checkup cohorts. Discrimination was evaluated using area under the receiver operating characteristic curve (AUC) and area under the precision-recall curve (AUPRC), and DLLV-based prescreening was evaluated by estimated TTE reduction and missed-case burden. Results The development dataset included 24,674 patients (32,135 CXR-TTE pairs); external cohorts included 5440, 28,034, and 6085 patients (53-62% male). DLLV achieved an AUC of 0.86 (95% CI: 0.84, 0.88) for LVEF ≤ 40% in the tertiary referral cohort. In health checkup cohorts, AUCs were 0.92-0.97 for LVEF ≤ 40%, with AUPRCs of 0.04 in both cohorts, corresponding to 87-fold (95% CI: 16, 404) and 37-fold (95% CI: 3, 251) enrichment over baseline prevalence. Counterfactual reinforcement learning produced heatmaps with greater heart localization precision (<i>P</i> < .001 for all) and higher AUC for LVEF < 50% (<i>P</i> = .04). DLLV-based prescreening estimated TTE volume reductions up to 85% while identifying all 12 and 7 individuals with LVEF ≤ 40%. Conclusion DLLV identified reduced LVEF on CXRs across external test cohorts and showed potential as a prescreening tool for selective TTE in health checkup populations. ©RSNA, 2026.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260161"},"PeriodicalIF":20.1,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149239695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Agentic Automation of BT-RADS Scoring Using an End-to-End System for Standardized Glioma Follow-up Assessment. 使用端到端系统标准化胶质瘤随访评估BT-RADS评分的代理自动化。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-10-21 DOI: 10.1148/ryai.260234
Mohamed Sobhi Jabal, Jikai Zhang, Dominic LaBella, Jessica L Houk, Dylan Zhang, Jeffrey D Rudie, Maciej A Mazurowski, Kirti Magudia, Evan Calabrese
{"title":"Agentic Automation of BT-RADS Scoring Using an End-to-End System for Standardized Glioma Follow-up Assessment.","authors":"Mohamed Sobhi Jabal, Jikai Zhang, Dominic LaBella, Jessica L Houk, Dylan Zhang, Jeffrey D Rudie, Maciej A Mazurowski, Kirti Magudia, Evan Calabrese","doi":"10.1148/ryai.260234","DOIUrl":"https://doi.org/10.1148/ryai.260234","url":null,"abstract":"<p><p>The Brain Tumor Reporting and Data System (BT-RADS) standardizes posttreatment MRI response assessment in patients with diffuse gliomas but requires complex integration of imaging trends, medication effects, and radiation timing. An end-to-end multiagent large language model and convolutional neural network system for automated BTRADS classification was developed and evaluated. Five hundred nine consecutive posttreatment glioma MRI examinations (March 2017-July 2021) from a single center were retrospectively evaluated. An extractor agent identified clinical variables from notes using gpt-oss (OpenAI, 20-billion parameters). A scorer agent applied BTRADS decision logic integrating these with quantitative volumetrics. System accuracy was compared against initial assessments, using expert annotation as the reference standard. Four hundred ninety-two examinations were evaluable. Overall system accuracy was 76.0% (374/492) versus 57.5% (283/492) for initial assessments, an absolute difference of +18.5 percentage points (<i>P</i> < .001). Cohen kappa between the system and the reference standard was 0.71. The system achieved high sensitivity in context-dependent categories (BT-1b; 100%; BT-1a, 92.7%; BT-3a, 87.5%) and moderate sensitivity for threshold-dependent categories (BT-3c, 74.8%; BT-2, 69.2%; BT-4, 69.3%; BT-3b, 57.1%). The BT-4 positive predictive value was 92.9%. These results support further evaluation of this system as a clinical decision-support tool, with local open-weight inference preserving data privacy.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260234"},"PeriodicalIF":20.1,"publicationDate":"2026-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149239697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Comparative Performance of Federated and Centralized Learning for Brain Tumor Classification and Segmentation. 联邦学习与集中式学习在脑肿瘤分类与分割中的比较性能。
IF 20.1
Radiology-Artificial Intelligence Pub Date : 2026-10-07 DOI: 10.1148/ryai.260515
Anish Narayan, Chun Goh, Frederick Mariajoseph, Shujie Cui, Enes Makalic, Justin Moore
{"title":"Comparative Performance of Federated and Centralized Learning for Brain Tumor Classification and Segmentation.","authors":"Anish Narayan, Chun Goh, Frederick Mariajoseph, Shujie Cui, Enes Makalic, Justin Moore","doi":"10.1148/ryai.260515","DOIUrl":"https://doi.org/10.1148/ryai.260515","url":null,"abstract":"<p><p>Purpose To compare federated learning (FL) and centralized learning (CL) performance for brain tumor classification and segmentation and examine whether specific configurations confer measurable performance advantages. Materials and Methods In this systematic review and meta-analysis, Medline, EMBASE, PubMed, Web of Science, and Scopus were searched from inception to March 4, 2026. Classification accuracy was pooled using random-effects meta-analysis, with heterogeneity quantified using I<sup>2</sup>; other metrics were summarized descriptively. Paired differences in classification accuracy and Dice similarity coefficient (DSC) were assessed using the Wilcoxon signed-rank test. Mixed-effects meta-regression and subgroup analyses evaluated whether specific configurations were associated with performance differences. Results Of 3,880 records identified, 50 studies were included (26 classification/detection, 26 segmentation; 2 in both). Within-study differences between FL and CL performance were -0.87 percentage points (pp) for classification accuracy (95% CI [-3.32, +1.44]; <i>P</i> = .41, k = 13) and -1.29 pp for segmentation DSC (95% CI [-4.23, +0.81]; Holm-corrected <i>P</i> = .50, k = 11). In unmatched single-arm analyses, pooled classification accuracy was 96.77% for FL (k = 18) versus 96.86% for CL (k = 9); the pooled DSC for segmentation was 84.95% (k = 24) versus 84.61% (k = 11). Heterogeneity was high (I<sup>2</sup>=93.3-96.5%); the 95% prediction interval for FL accuracy was 86.72-100%. Aggregation strategy, training sample size, task category, and independent and identically distributed (IID) versus non-IID partitioning were not associated with performance differences (all <i>P</i> ≥ .38). Conclusion There was no evidence of paired differences between FL and CL in brain tumor classification accuracy or segmentation DSC, and no specific configuration conferred a measurable performance advantage. © RSNA, 2026.</p>","PeriodicalId":29787,"journal":{"name":"Radiology-Artificial Intelligence","volume":" ","pages":"e260515"},"PeriodicalIF":20.1,"publicationDate":"2026-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149128609","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书