Journal of the American College of Radiology最新文献

{"title":"A Learning Accelerator Framework: Scalable Clinical Artificial Intelligence Development and Delivery","authors":"Diana S.M. Buist PhD ,&nbsp;Annie Y. Ng PhD ,&nbsp;Bryan Haslam PhD ,&nbsp;Edgar A. Wakelin PhD ,&nbsp;Christoph I. Lee MD, MS, MBA ,&nbsp;Sham Sokka PhD ,&nbsp;A. Gregory Sorensen MD","doi":"10.1016/j.jacr.2025.12.015","DOIUrl":"10.1016/j.jacr.2025.12.015","url":null,"abstract":"<div><h3>Objectives</h3><div>To introduce a vertically integrated model between a health care service provider and technology developer as a learning accelerator to address challenges in developing and delivering artificial intelligence (AI) into health care.</div></div><div><h3>Methods</h3><div>The Learning Accelerator Framework is built on four core components that focus on improving patient and health care outcomes: an integrated data registry, a continuous technology development stack, adaptive clinical services, and an iterative learning and development loop. Its application is described in one case study to highlight its operational mechanisms throughout the AI life cycle.</div></div><div><h3>Results</h3><div>The framework has guided the conceptualization, development, implementation, and national delivery of a multistage AI breast cancer screening workflow, progressing from initial clinical validation (thousands) to population-scale implementation (millions of patients). We demonstrate how iterative learning loops were applied using clinical feedback and real-world data monitoring feedback, which resulted in a multistage AI screening workflow that has achieved a significant absolute increase in cancer detection rate (Δ0.99 cancers per 1,000 examinations [95% confidence interval: 0.59-1.42]) and positive predictive value (Δ0.55 cancers per 100 recalls [95% confidence interval: 0.30-1.03]) with equitable benefits across breast density, race, and ethnic subpopulations.</div></div><div><h3>Discussion</h3><div>The Learning Accelerator Framework represents a departure from traditional approaches by mitigating challenges, inefficiencies, and delays that impede AI translation, offering a model for AI developers and provider systems seeking to accelerate innovation. The breast AI case study demonstrates how instrumental the framework can be for ensuring ongoing AI implementation effectiveness, fostering clinician trust, and ultimately improving operations, patient outcomes and health equity.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 809-817"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145758320","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recommendations From the Blue Ribbon Panel on Fluoroscopy Safety","authors":"Dustin A. Gress MS ,&nbsp;M. Mahesh MS, PhD ,&nbsp;Kevin W. Dickey MD ,&nbsp;John F. Angle MD ,&nbsp;D. Duane Baldwin MD ,&nbsp;Stephen Balter PhD ,&nbsp;Wayne Batchelor MD, MHS, MBA ,&nbsp;Lisa Bruedigan ,&nbsp;Christopher Davis DMSc, PA-C, RT ,&nbsp;Deirdre Elder MS ,&nbsp;R. Paul Guillerman MD ,&nbsp;Maged N. Guirguis MD ,&nbsp;David Hardwick MSRS, RRA, RT(R) ,&nbsp;Carrie M. Hayes DMSc, PA-C, RDMS, RVT ,&nbsp;Jeremy J. Heit MD, PhD ,&nbsp;A. Kyle Jones PhD ,&nbsp;Melissa Kirkwood MD ,&nbsp;Andrew Kuhls-Gilcrist PhD ,&nbsp;Bonnie Martin-Harris PhD ,&nbsp;William W. Mayo-Smith MD ,&nbsp;Alan H. Matsumoto MD","doi":"10.1016/j.jacr.2025.12.020","DOIUrl":"10.1016/j.jacr.2025.12.020","url":null,"abstract":"<div><div>There are many challenges associated with the safe use of fluoroscopy. These challenges include but are not limited to highly variable regulatory requirements, scope of practice concerns, inconsistent education and training, and lack of staff empowerment. Challenges are further compounded by the increasing use of fluoroscopy across a wide range of medical specialties. To facilitate consensus on how to address the issues, the ACR convened the multidisciplinary Blue Ribbon Panel on Fluoroscopy Safety (BRP-FS), with 32 organizations represented. The goal of the BRP-FS is to establish multi- and interspecialty consensus standards for the safe use of fluoroscopy in health care, including minimum and uniform standards for the education and training of fluoroscopy users that apply across geographic and professional boundaries, for the benefit of all patients and health care providers. Recommendations are made for local practices, professional organizations, industry, regulatory agencies, and accreditation bodies. Foundational to the recommendations of the BRP-FS are the personnel training and procedure classification frameworks in National Council on Radiation Protection and Measurement Commentary No. 33.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 779-789"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146133813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ACR Appropriateness Criteria® Nonthrombotic Iliac Vein Lesion","authors":"Expert Panel on Vascular Imaging,&nbsp;Nicole A. Keefe MD ,&nbsp;Sandra Gad MD ,&nbsp;Minhaj S. Khaja MD, MBA ,&nbsp;Nima Kokabi MD ,&nbsp;Anant D. Bhave MD ,&nbsp;Umberto Campia MD ,&nbsp;Benjamin N. Contrella MD ,&nbsp;Saeed Elojeimy MD, PhD ,&nbsp;Edward Hulten MD, MPH ,&nbsp;Baljendra S. Kapoor MD, MBA ,&nbsp;Michael Malinowski MD ,&nbsp;Mark Meissner MD ,&nbsp;Rachel P. Rosovsky MD ,&nbsp;Fadi Shamoun MD ,&nbsp;Aditya M. Sharma MBBS ,&nbsp;Daniel P. Sheeran MD ,&nbsp;Bill S. Majdalany MD","doi":"10.1016/j.jacr.2026.01.021","DOIUrl":"10.1016/j.jacr.2026.01.021","url":null,"abstract":"<div><div>Nonthrombotic iliac vein lesions (NIVLs), also referred to as a May-Thurner lesion/anatomy or Cockett syndrome, most frequently result from the left common iliac vein being compressed between the right common iliac artery and the spine. Iliac vein compression is often clinically silent, but can be associated with significant symptoms in which case it is referred to as May-Thurner syndrome. It is still unclear why some patients are asymptomatic whereas others can develop severe, debilitating symptoms. Most frequently, this disorder can cause lower extremity edema, deep venous thrombosis, or chronic venous insufficiency. Since NIVL can be clinically silent, a thorough history and physical examination is needed to rule out alternative etiologies. For symptomatic patients, imaging evaluation is warranted to make a definitive diagnosis. The objectives of imaging in a symptomatic patient include confirmation of diagnosis, identifying the location and extent of occlusion, and procedural planning.</div><div>The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 887-906"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146215191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Imaging Utilization Among Pediatric Beneficiaries at Children’s Hospital Versus Non–Children’s Hospital Outpatient Facilities Using Medicaid Claims","authors":"Casey E. Pelzl MPH ,&nbsp;Andrea S. Doria MD ,&nbsp;Alexandra Drake MPH ,&nbsp;Eric W. Christensen PhD ,&nbsp;Michael S. Gee MD ,&nbsp;Shireen E. Hayatghaibi PhD ,&nbsp;Mai-Lan Ho MD ,&nbsp;Sara R. Teixeira MD ,&nbsp;Andrew T. Trout MD ,&nbsp;Elysa Widjaja MD ,&nbsp;Elizabeth Y. Rula PhD ,&nbsp;Sherwin S. Chan MD","doi":"10.1016/j.jacr.2025.11.029","DOIUrl":"10.1016/j.jacr.2025.11.029","url":null,"abstract":"<div><h3>Purpose</h3><div>The aim of this study was to compare imaging use on pediatric outpatients at children’s hospitals (CHs) versus non–children’s hospitals (NCHs) to identify differences across modalities that differ in ionizing radiation exposure.</div></div><div><h3>Methods</h3><div>CMS Medicaid Research Identifiable Files were used to identify all year 2019 pediatric (ages 0-17 years) outpatient claims from hospital outpatient facilities (HOFs) and emergency departments (EDs). CMS data from 2018 were used to calculate the pediatric comorbidity index (PCI) for risk adjustment. Primary outcomes were CT, MR, ultrasound, or radiography (XR) use at each visit, comparing frequencies between CHs and NCHs. Additional covariates included age group (0, 1-2, 3-5, 6-11, and 12-17 years), PCI (0, 1 or 2, 3-6, 7), and place of service (HOF vs ED).</div></div><div><h3>Results</h3><div>A total of 5,474,082 claims meeting the selection criteria were identified. More than half of visits (53%) were to CHs, and 15% were to EDs. CH encounters were more likely (vs NCH encounters) to be among patients aged 0 to 5 years versus &gt;5 years (41.2% vs 38.7%, <em>P</em> &lt; .01), those with PCI &gt; 2 (32.3% vs 22.9%, <em>P</em> &lt; .01), and those seen at HOFs (87.8% vs 81.9%, <em>P</em> &lt; .01). The most commonly used modalities were XR (9.5%) and ultrasound (2.1%). Use of XR (11.8% vs 7.5%, <em>P</em> &lt; .01) and CT (1.0% vs 0.5%, <em>P</em> &lt; .01) was more frequent at NCHS. Use of ultrasound (2.5% vs 1.7%, <em>P</em> &lt; .01) and MR (0.9% vs 0.5%, <em>P</em> &lt; .01) was more frequent at CHs.</div></div><div><h3>Conclusions</h3><div>This study reveals that imaging modalities that expose children to ionizing radiation are used more frequently at NCHs than at CHs. The clinical implications of these variations warrant further investigation.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 757-767"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146013531","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating Generative Artificial Intelligence as an Educational Tool for Radiology Resident Report Drafting","authors":"Antonio Verdone MSc ,&nbsp;Aidan Cardall BS ,&nbsp;Fardeen Siddiqui BS ,&nbsp;Motaz Nashawaty MD ,&nbsp;Danielle Rigau MD ,&nbsp;Youngjoon Kwon MD ,&nbsp;Mira Yousef MD ,&nbsp;Shalin Patel MD ,&nbsp;Alex Kieturakis MD ,&nbsp;Eric Kim MD ,&nbsp;Laura Heacock MD ,&nbsp;Beatriu Reig MD ,&nbsp;Yiqiu Shen PhD","doi":"10.1016/j.jacr.2025.12.024","DOIUrl":"10.1016/j.jacr.2025.12.024","url":null,"abstract":"<div><h3>Objective</h3><div>Radiology residents require timely, personalized feedback to develop accurate image analysis and reporting skills. Increasing clinical workload often limits attendings’ ability to provide guidance. This study evaluates a HIPAA-compliant Generative Pretrained Transformer (GPT)-4o system that delivers automated feedback on breast imaging reports drafted by residents in real clinical settings.</div></div><div><h3>Methods</h3><div>We analyzed 5,000 resident-attending report pairs from routine practice at a multisite US health system. GPT-4o was prompted with clinical instructions to identify common errors and provide feedback. A reader study using 100 report pairs was conducted. Four attending radiologists and four residents independently reviewed each pair, determined whether predefined error types were present, and rated GPT-4o’s feedback as helpful or not. Agreement between GPT and readers was assessed using percent match. Interreader reliability was measured with Krippendorff’s α. Educational value was measured as the proportion of cases rated helpful.</div></div><div><h3>Results</h3><div>Three common error types were identified: (1) omission or addition of key findings, (2) incorrect use or omission of technical descriptors, and (3) final assessment inconsistent with findings. GPT-4o showed strong agreement with attending consensus: 90.5%, 78.3%, and 90.4% (Cohen’s κ: 0.790, 0.550, and 0.615) across error types. Interreader reliability among all eight readers showed moderate to substantial variability (α = 0.767, 0.595, 0.567). When each reader was individually replaced with GPT-4o and interreader agreement among seven readers and GPT was recalculated, the effect was not statistically significant (Δ = −0.004 to 0.002, all <em>P</em> &gt; .05). GPT’s feedback was rated helpful in most cases: 89.8%, 83.0%, and 92.0%.</div></div><div><h3>Discussion</h3><div>ChatGPT-4o can reliably identify key educational errors. It may serve as a scalable tool to support radiology education.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 818-826"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145844501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dr No: The Art of Saying “No” (and Feeling Good About It)","authors":"Kirang Patel MD ,&nbsp;Yasha Gupta MD ,&nbsp;Alex Podlaski MD","doi":"10.1016/j.jacr.2026.01.016","DOIUrl":"10.1016/j.jacr.2026.01.016","url":null,"abstract":"","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 856-858"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146100965","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Differences in Pediatric Imaging Utilization Between Children’s and Non–Children’s Hospital: Is It Time to Move On From the Focus on CT and Radiation?","authors":"Hansel J. Otero MD ,&nbsp;Taisa Guarilha MD","doi":"10.1016/j.jacr.2026.01.011","DOIUrl":"10.1016/j.jacr.2026.01.011","url":null,"abstract":"","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 768-769"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146088154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiology Board-Style Examinations and Large Language Models: A Scoping Review of Model Performance","authors":"Pilar López-Úbeda PhD ,&nbsp;Teodoro Martín-Noguerol MD ,&nbsp;Antonio Luna MD, PhD","doi":"10.1016/j.jacr.2026.01.017","DOIUrl":"10.1016/j.jacr.2026.01.017","url":null,"abstract":"<div><h3>Background</h3><div>Large language models (LLMs) are increasingly being evaluated for their ability to answer official radiology board-style examination questions. Understanding their accuracy, limitations, and potential applications in education is essential for assessing their utility in the field.</div></div><div><h3>Material and methods</h3><div>A scoping review was conducted in October 2025 across PubMed, Scopus, and Web of Science, following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Studies were included if they evaluated LLMs on official radiology board-style examination questions. After screening 205 unique records, 29 studies met the inclusion criteria. Data were extracted on study characteristics, including LLM type and version, input modality, language, examination type, answer format, comparison with humans, and reported outcomes.</div></div><div><h3>Results</h3><div>The reviewed studies evaluated multiple LLMs, predominantly Chat Generative Pre-trained Transformer (GPT)-based models (GPT-3.5, GPT-4, GPT-4 Turbo, GPT-4o), as well as Claude, Gemini, Llama 3, and Mixtral. Text-only evaluations generally yielded higher accuracy (≈65%-90%) compared with multimodal tasks (45%-89%). GPT-4 and its variants consistently outperformed earlier versions, occasionally exceeding average human performance. Open-source models such as Llama 3 70B and Mixtral achieved comparable results to proprietary models, offering advantages in local deployment and privacy. Few studies directly compared LLM performance with human radiologists.</div></div><div><h3>Conclusions</h3><div>LLMs demonstrate promising performance in answering text-based radiology board-style examination questions, particularly GPT-4-based models. Nevertheless, significant limitations persist in multimodal tasks and complex reasoning scenarios.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 837-848"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146094985","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ACR Appropriateness Criteria® Imaging After Liver Transplant","authors":"Expert Panel on Gastrointestinal Imaging,&nbsp;Shaun A. Wahab MD ,&nbsp;Abhinav Vij MD, MPH ,&nbsp;Alice Fung MD ,&nbsp;Mustafa R. Bashir MD ,&nbsp;Brooks D. Cash MD ,&nbsp;Elizabeth M. Hecht MD ,&nbsp;A. Tuba Karagulle Kendi MD ,&nbsp;Brendan M. McGuire MD ,&nbsp;Gregory K. Russo MD ,&nbsp;Elainea N. Smith MD ,&nbsp;Kiran H. Thakrar MD ,&nbsp;Lisa B. VanWagner MD, MSc ,&nbsp;Atif Zaheer MD ,&nbsp;Kathryn J. Fowler MD","doi":"10.1016/j.jacr.2026.01.023","DOIUrl":"10.1016/j.jacr.2026.01.023","url":null,"abstract":"<div><div>Liver transplantation is currently the treatment of choice for patients with acute or advanced chronic liver failure. Complications that can lead to liver allograft failure or patient mortality include vascular abnormalities, biliary complications, infection, rejection, and recurrent or posttransplant malignancy. Imaging plays a vital role in detecting these complications. This document summarizes the relevant literature for the selection of imaging after liver transplant for the following clinical scenarios: immediate postoperative period, postoperative complications of suspected vascular etiology, postoperative complications of suspected biliary etiology, and postoperative surveillance.</div><div>The American College of Radiology Appropriateness Criteria are evidence-based guidelines for specific clinical conditions that are reviewed annually by a multidisciplinary expert panel. The guideline development and revision process support the systematic analysis of the medical literature from peer reviewed journals. Established methodology principles such as Grading of Recommendations Assessment, Development, and Evaluation or GRADE are adapted to evaluate the evidence. The RAND/UCLA Appropriateness Method User Manual provides the methodology to determine the appropriateness of imaging and treatment procedures for specific clinical scenarios. In those instances where peer reviewed literature is lacking or equivocal, experts may be the primary evidentiary source available to formulate a recommendation.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 859-873"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146215138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"More Than One Path: Recent Cross-Application Trends Among Radiology Residency Applicants, 2020-2023","authors":"Nauman Hussain BS ,&nbsp;Angela Renne BS ,&nbsp;Erin Gomez MD ,&nbsp;Claire Brookmeyer MD ,&nbsp;Jenny X. Chen MD, EdM ,&nbsp;Francis Deng MD","doi":"10.1016/j.jacr.2026.01.010","DOIUrl":"10.1016/j.jacr.2026.01.010","url":null,"abstract":"<div><h3>Objective</h3><div>Cross-application is increasingly common in radiology yet remains understudied. We aimed to quantify the prevalence, characterize patterns, and identify predictors of cross-application among diagnostic radiology (DR) and interventional radiology (IR) applicants.</div></div><div><h3>Methods</h3><div>We analyzed all Electronic Residency Application Service applications submitted to DR and IR programs between 2020 and 2023. Cross-applications were defined as those applying to DR or IR and at least one nonradiology specialty. Applicant demographics, educational background, and academic characteristics were compared using χ² and <em>t</em> tests. Independent predictors of cross-application were assessed using multivariable logistic regression.</div></div><div><h3>Results</h3><div>Among 9,012 radiology applications, 41.8% (n = 3,770) were cross-applications. Cross-application was most frequent among IR-only applications (94.0%) and more common in international medical graduates (63.9%), DOs (46.6%), and applicants ≥30 years old (56.0%). Compared with radiology-only applications, cross-applications had more publications but fewer research and volunteer experiences. Independent predictors of cross-application included IR-only status (odds ratio [OR] = 31.75; 95% confidence interval [CI]: 22.42-46.54), international medical graduate (OR = 2.86; 95% CI: 2.47-3.31), DO (OR = 1.90; 95% CI: 1.65-2.18), age ≥ 30 (OR = 1.71; 95% CI: 1.54-1.90), and re-application (OR = 2.10; 95% CI: 1.80-2.44), and Alpha Omega Alpha society membership reduced odds (OR = 0.48; 95% CI: 0.39-0.59).</div></div><div><h3>Conclusion</h3><div>Cross-application is a frequent strategy among radiology applicants. These results highlight the challenges of navigating a competitive match and the importance of advising that accounts for diverse applicant backgrounds and career goals. Understanding cross-application as a common and often pragmatic strategy may help inform advising practices and reduce misinterpretation of applicant intent.</div></div>","PeriodicalId":49044,"journal":{"name":"Journal of the American College of Radiology","volume":"23 5","pages":"Pages 827-833"},"PeriodicalIF":5.1,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146055338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}