Journal of the American College of Radiology : JACR最新文献

{"title":"Using a Large Language Model for Postdeployment Monitoring of FDA-Approved Artificial Intelligence: Pulmonary Embolism Detection Use Case.","authors":"Vera Sorin, Panagiotis Korfiatis, Alex K Bratt, Tim Leiner, Christoph Wald, Crystal Butler, Cole J Cook, Timothy L Kline, Jeremy D Collins","doi":"10.1016/j.jacr.2025.06.036","DOIUrl":"10.1016/j.jacr.2025.06.036","url":null,"abstract":"<p><strong>Background: </strong>Artificial intelligence (AI) is increasingly integrated into clinical workflows. The performance of AI in production can diverge from initial evaluations. Postdeployment monitoring (PDM) remains a challenging ingredient of ongoing quality assurance once AI is deployed in clinical production.</p><p><strong>Purpose: </strong>To develop and evaluate a PDM framework that uses large language models (LLMs) for free-text classification of radiology reports, and human oversight. We demonstrate its application to monitor a commercially vended pulmonary embolism (PE) detection AI (CVPED).</p><p><strong>Methods: </strong>We retrospectively analyzed 11,999 CT pulmonary angiography studies performed between April 30, 2023, and June 17, 2024. Ground truth was determined by combining LLM-based radiology report classification and the CVPED outputs, with human review of discrepancies. We simulated a daily monitoring framework to track discrepancies between CVPED and the LLM. Drift was defined when discrepancy rate exceeded a fixed 95% confidence interval for 7 consecutive days. The confidence interval and the optimal retrospective assessment period were determined from a stable dataset with consistent performance. We simulated drift by systematically altering CVPED or LLM sensitivity and specificity, and we modeled an approach to detect data shifts. We incorporated a human-in-the-loop selective alerting framework for continuous prospective evaluation and to investigate potential for incremental detection.</p><p><strong>Results: </strong>Of 11,999 CT pulmonary angiography studies, 1,285 (10.7%) had PE. Overall, 373 (3.1%) had discrepant classifications between CVPED and LLM. Among 111 CVPED-positive and LLM-negative cases, 29 would have triggered an alert due to the radiologist not interacting with CVPED. Of those, 24 were CVPED false-positives, 1 was an LLM false-negative, and the framework ultimately identified 4 true-alerts for incremental PE cases. The optimal retrospective assessment period for drift detection was determined to be 2 months. A 2% to 3% decline in model specificity caused a 2- to 3-fold increase in discrepancies, and a 10% drop in sensitivity was required to produce a similar effect. For example, a 2.5% drop in LLM specificity led to a 1.7-fold increase in CVPED-negative LLM-positive discrepancies, which would have taken 22 days to detect using the proposed framework.</p><p><strong>Conclusion: </strong>A PDM framework combining LLM-based free-text classification with a human-in-the-loop alerting system can continuously track an image-based AI's performance, alert for performance drift, and provide incremental clinical value.</p>","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144556047","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}

{"title":"Closing the Loop: From Global Screening to On-Time Access.","authors":"Javier Alejandro Lamprea Ardila, José David Cardona Ortegón, Laura Manuela Olarte Bermúdez, Javier Andrés Romero","doi":"10.1016/j.jacr.2025.06.039","DOIUrl":"10.1016/j.jacr.2025.06.039","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144531484","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}

{"title":"Radiology Readouts: Faculty and Trainee Perceptions and Preferences of the Current State.","authors":"Shanna A Matalon, Sophia R O'Brien, Jeffrey P Guenette, Scott Simpson","doi":"10.1016/j.jacr.2025.06.035","DOIUrl":"10.1016/j.jacr.2025.06.035","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144531486","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}

{"title":"Interventional Radiology and Health Equity: Disparities in Uterine Artery Embolization Based on Insurance Status.","authors":"Brijesh Sathian, Syed Muhammad Ali, Javed Iqbal, Ayesha Parvaiz Malik","doi":"10.1016/j.jacr.2025.06.034","DOIUrl":"10.1016/j.jacr.2025.06.034","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144531485","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}

{"title":"Agentic AI: The Power to Change Medicine and Our World.","authors":"Kimberly Powell, Elliot K Fishman, Linda C Chu, Steven P Rowe, Charles K Crawford","doi":"10.1016/j.jacr.2025.06.032","DOIUrl":"https://doi.org/10.1016/j.jacr.2025.06.032","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144512903","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}

{"title":"A Review of Cybersecurity and Privacy Standards in Medical Imaging: Consensus, Frameworks, and Best Practices.","authors":"Po-Hao Chen, Christoph Wald","doi":"10.1016/j.jacr.2025.06.033","DOIUrl":"10.1016/j.jacr.2025.06.033","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144512902","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}

{"title":"State of the Field: Radiology Residency Clinician Educator Tracks.","authors":"Sophia R O'Brien, Erin Gomez, Kirang Patel, Shanna Matalon, Mohamed S Muneer, Alex Le Lindqwister, Rowa A Mohamed, Margaret Lin, Scott Simpson","doi":"10.1016/j.jacr.2025.06.031","DOIUrl":"10.1016/j.jacr.2025.06.031","url":null,"abstract":"<p><strong>Objective: </strong>Radiology resident clinician educator tracks (CETs) are designed to prepare residents for a career in academia. The number and structure of US radiology CETs is unknown. This study sought to describe the current state of the field of US diagnostic radiology CETs.</p><p><strong>Methods: </strong>This multimethod observational cross-sectional study involved an online survey and virtual semistructured interviews. AMA Freida website search was performed in February 2024 to identify programs with radiology CETs. All identified tracks were sent the online survey, and three programs with a known track or track in creation not listed in AMA Freida were also surveyed. Half of the surveyed CETs were randomly invited to interview.</p><p><strong>Results: </strong>A total of 23 active or potentially active radiology CETs were identified, reflecting 11.6% of all US diagnostic radiology residencies, of which 18 programs responded (78% response rate). CET length and structure varied. A slight majority of tracks are resident-led (9 of 17, 53%). Most faculty track leaders do not receive protected academic time for their role (15 of 17, 88%), and only half of the CETs reported sufficient institutional support. The most common key components of radiology CETs are lectures, track meetings, mentorship, and capstone projects. All CETs directly observe resident teaching. Four of 9 invited programs (44%) completed semistructured interviews and described the importance of, and challenges with, faculty engagement and resident project completion.</p><p><strong>Discussion: </strong>This study describes the state of radiology CETs, including their keys to success, challenges, and recommendations for others creating a track. Future research should investigate CET outcomes and further explore resident experiences.</p>","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144478076","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}

{"title":"Environmental Impact of Inappropriate Imaging: Strengthening the Call for Sustainable Radiology Practice.","authors":"Brijesh Sathian, Javed Iqbal, Syed Muhammad Ali","doi":"10.1016/j.jacr.2025.06.019","DOIUrl":"10.1016/j.jacr.2025.06.019","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144369627","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}

{"title":"Addressing the Screening Care Gap by Removing Upfront Fees for Digital Breast Tomosynthesis.","authors":"Matthew D Phelps, Brian N Dontchos","doi":"10.1016/j.jacr.2025.05.024","DOIUrl":"10.1016/j.jacr.2025.05.024","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144369626","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}

{"title":"Pearls and Pitfalls of Integrating Artificial Intelligence Into Radiology Education.","authors":"Priscilla J Slanetz, Wasif Bala, Florence X Doo, Tessa Cook","doi":"10.1016/j.jacr.2025.06.026","DOIUrl":"10.1016/j.jacr.2025.06.026","url":null,"abstract":"","PeriodicalId":73968,"journal":{"name":"Journal of the American College of Radiology : JACR","volume":" ","pages":""},"PeriodicalIF":0.0,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144340767","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}