{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T11:04:01Z","timestamp":1790075041581,"version":"4.0.1"},"reference-count":42,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T00:00:00Z","timestamp":1790035200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T00:00:00Z","timestamp":1790035200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Beijing Natural Science Foundation","award":["7254539"],"award-info":[{"award-number":["7254539"]}]},{"name":"Beijing Natural Science Foundation","award":["L242024"],"award-info":[{"award-number":["L242024"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62522119"],"award-info":[{"award-number":["62522119"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100019398","name":"Beijing Friendship Hospital, Capital Medical University","doi-asserted-by":"publisher","award":["YYZZ202334"],"award-info":[{"award-number":["YYZZ202334"]}],"id":[{"id":"10.13039\/501100019398","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Open Projects of Sichuan Province Clinical Medical Research Center for Imaging Medicine","award":["YXYX2409"],"award-info":[{"award-number":["YXYX2409"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Eur Radiol Exp"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Errors in radiology reports are a major patient-safety concern and are difficult to detect with manual quality assurance (QA). Large language models (LLMs) can assist, but generic prompting does not reflect radiologists\u2019 structured, section-based workflows.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To develop and evaluate RadCoT (Radiological Chain-of-Thought), a domain-specific prompting framework aligning LLM reasoning with radiological review workflows, and to assess whether it enables open-source models to approach commercial benchmarks for error detection.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>In this retrospective study, 1,170 clinician-validated errors were extracted from a departmental QA repository (January 2021\u2013December 2024), corresponding to 900 error-containing reports. An additional 300 error-free reports served as controls, yielding 1,200 reports balanced across radiography, ultrasound, CT, and MRI. Errors were categorized into five types by experienced radiologists. Seven LLMs were evaluated using standard prompting and the six-step RadCoT framework. Micro-averaged precision, recall, and F1 were computed at the error-instance level. Error type, modality-specific performance, and inference time were analyzed.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      RadCoT significantly improved the mean micro-averaged F1 across all models from 0.77\u2009\u00b1\u20090.06 (standard) to 0.85\u2009\u00b1\u20090.06 (RadCoT;\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.003). GPT-4o with RadCoT achieved the highest F1 (0.93). Llama-3.3-70B with RadCoT (F1\u2009=\u20090.89) narrowed the gap with GPT-4o under standard prompting (F1\u2009=\u20090.88; paired\n                      <jats:italic>t<\/jats:italic>\n                      -test, Holm-adjusted\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.28). Interpretation errors showed the largest gain, with F1 improving from 0.57 to 0.75.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>RadCoT consistently enhances LLM-based error detection, particularly for complex logic and consistency errors, closing the gap between open-source and commercial models and offering a pathway for privacy-preserving, on-premises QA.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Key Points<\/jats:title>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Question<\/jats:italic>\n                      <\/jats:bold>\n                      Manual review misses radiology report errors and remains difficult to scale.\n                    <\/jats:p>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Findings<\/jats:italic>\n                      <\/jats:bold>\n                      Structured prompting improves detection of interpretation and section-consistency errors across seven models.\n                    <\/jats:p>\n                    <jats:p>\n                      <jats:bold>\n                        <jats:italic>Relevance statement<\/jats:italic>\n                      <\/jats:bold>\n                      Open-source models approach commercial performance for privacy-preserving on-premises radiology quality assurance.\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s41747-026-00804-0","type":"journal-article","created":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T10:11:40Z","timestamp":1790071900000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["RadCoT: a Radiological Chain-of-Thought framework for enhanced error detection in radiology reports"],"prefix":"10.1186","volume":"10","author":[{"given":"Jia","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zichun","family":"Zhou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yantao","family":"Niu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Pengfei","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yan","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinghao","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lihua","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lining","family":"Dong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Wei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuan","family":"Wei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenchang","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9559-4777","authenticated-orcid":false,"given":"Han","family":"Lv","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,9,22]]},"reference":[{"key":"804_CR1","doi-asserted-by":"publisher","first-page":"171","DOI":"10.1007\/s13244-016-0534-1","volume":"8","author":"AP Brady","year":"2017","unstructured":"Brady AP (2017) Error and discrepancy in radiology: inevitable or avoidable? Insights Imaging 8:171\u2013182. https:\/\/doi.org\/10.1007\/s13244-016-0534-1","journal-title":"Insights Imaging"},{"key":"804_CR2","doi-asserted-by":"publisher","first-page":"739","DOI":"10.2214\/AJR.16.16963","volume":"208","author":"S Waite","year":"2017","unstructured":"Waite S, Scott J, Gale B, Fuchs T, Kolla S, Reede D (2017) Interpretive error in radiology. AJR Am J Roentgenol 208:739\u2013749. https:\/\/doi.org\/10.2214\/AJR.16.16963","journal-title":"AJR Am J Roentgenol"},{"key":"804_CR3","doi-asserted-by":"publisher","first-page":"377","DOI":"10.4329\/wjr.v2.i10.377","volume":"2","author":"A Pinto","year":"2010","unstructured":"Pinto A, Brunese L (2010) Spectrum of diagnostic errors in radiology. World J Radiol 2:377\u2013383. https:\/\/doi.org\/10.4329\/wjr.v2.i10.377","journal-title":"World J Radiol"},{"key":"804_CR4","doi-asserted-by":"publisher","first-page":"1668","DOI":"10.1148\/rg.2015150023","volume":"35","author":"MA Bruno","year":"2015","unstructured":"Bruno MA, Walker EA, Abujudeh HH (2015) Understanding and confronting our mistakes: the epidemiology of error in radiology and strategies for error reduction. Radiographics 35:1668\u20131676. https:\/\/doi.org\/10.1148\/rg.2015150023","journal-title":"Radiographics"},{"key":"804_CR5","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1148\/rg.292085090","volume":"29","author":"JB Kruskal","year":"2009","unstructured":"Kruskal JB, Anderson S, Yam CS, Sosna J (2009) Strategies for establishing a comprehensive quality and performance improvement program in a radiology department. Radiographics 29:315\u2013329. https:\/\/doi.org\/10.1148\/rg.292085090","journal-title":"Radiographics"},{"key":"804_CR6","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1038\/s44401-025-00023-6","volume":"2","author":"AB Jing","year":"2025","unstructured":"Jing AB, Garg N, Zhang J, Brown JJ (2025) AI solutions to the radiology workforce shortage. NPJ Health Syst 2:20. https:\/\/doi.org\/10.1038\/s44401-025-00023-6","journal-title":"NPJ Health Syst"},{"key":"804_CR7","doi-asserted-by":"publisher","first-page":"129","DOI":"10.1016\/j.jacr.2024.08.030","volume":"22","author":"B Siewert","year":"2025","unstructured":"Siewert B, Bruno MA, Bourland JD et al (2025) Seven challenges in radiology practice: from declining reimbursement to inadequate labor force: summary of the 2023 ACR Intersociety Meeting. J Am Coll Radiol 22:129\u2013138. https:\/\/doi.org\/10.1016\/j.jacr.2024.08.030","journal-title":"J Am Coll Radiol"},{"key":"804_CR8","doi-asserted-by":"publisher","first-page":"93","DOI":"10.1007\/s13244-011-0066-7","volume":"2","author":"European Society of Radiology (ESR)","year":"2011","unstructured":"European Society of Radiology (ESR) (2011) Good practice for radiological reporting. Guidelines from the European Society of Radiology (ESR). Insights Imaging 2:93\u201396. https:\/\/doi.org\/10.1007\/s13244-011-0066-7.","journal-title":"Insights Imaging"},{"key":"804_CR9","doi-asserted-by":"publisher","first-page":"852","DOI":"10.1148\/radiol.2523081992","volume":"252","author":"CE Kahn","year":"2009","unstructured":"Kahn CE, Langlotz CP, Burnside ES et al (2009) Toward best practices in radiology reporting. Radiology 252:852\u2013856. https:\/\/doi.org\/10.1148\/radiol.2523081992","journal-title":"Radiology"},{"key":"804_CR10","doi-asserted-by":"publisher","first-page":"184","DOI":"10.1148\/radiol.10101045","volume":"259","author":"JML Bosmans","year":"2011","unstructured":"Bosmans JML, Weyler JJ, De Schepper AM, Parizel PM (2011) The radiology report as seen by radiologists and referring clinicians: results of the COVER and ROVER surveys. Radiology 259:184\u2013195. https:\/\/doi.org\/10.1148\/radiol.10101045","journal-title":"Radiology"},{"key":"804_CR11","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1007\/s10278-012-9455-9","volume":"25","author":"CM Hawkins","year":"2012","unstructured":"Hawkins CM, Hall S, Hardin J, Salisbury S, Towbin AJ (2012) Prepopulated radiology report templates: a prospective analysis of error rate and turnaround time. J Digit Imaging 25:504\u2013511. https:\/\/doi.org\/10.1007\/s10278-012-9455-9","journal-title":"J Digit Imaging"},{"key":"804_CR12","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.230725","volume":"307","author":"LC Adams","year":"2023","unstructured":"Adams LC, Truhn D, Busch F et al (2023) Leveraging GPT-4 for post hoc transformation of free-text radiology reports into structured reporting: a multilingual feasibility study. Radiology 307:e230725. https:\/\/doi.org\/10.1148\/radiol.230725","journal-title":"Radiology"},{"key":"804_CR13","doi-asserted-by":"publisher","first-page":"4549","DOI":"10.1007\/s00330-025-11438-y","volume":"35","author":"B Salam","year":"2025","unstructured":"Salam B, St\u00fcwe C, Nowak S et al (2025) Large language models for error detection in radiology reports: a comparative analysis between closed-source and privacy-compliant open-source models. Eur Radiol 35:4549\u20134557. https:\/\/doi.org\/10.1007\/s00330-025-11438-y","journal-title":"Eur Radiol"},{"key":"804_CR14","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.232714","volume":"311","author":"RJ Gertz","year":"2024","unstructured":"Gertz RJ, Dratsch T, Bunck AC et al (2024) Potential of GPT-4 for detecting errors in radiology reports: implications for reporting accuracy. Radiology 311:e232714. https:\/\/doi.org\/10.1148\/radiol.232714","journal-title":"Radiology"},{"key":"804_CR15","unstructured":"Yu F, Endo M, Krishnan R et al (2023) Radiology Report Expert Evaluation (ReXVal) Dataset, version 1.0.0. PhysioNet. Available via https:\/\/physionet.org\/content\/rexval-dataset\/1.0.0\/. Accessed 11 Jun 2026"},{"key":"804_CR16","doi-asserted-by":"publisher","first-page":"2837","DOI":"10.1007\/s00330-021-08327-5","volume":"32","author":"JM Nobel","year":"2022","unstructured":"Nobel JM, van Geel K, Robben SGF (2022) Structured reporting in radiology: a systematic review to explore its potential. Eur Radiol 32:2837\u20132854. https:\/\/doi.org\/10.1007\/s00330-021-08327-5","journal-title":"Eur Radiol"},{"key":"804_CR17","doi-asserted-by":"publisher","first-page":"199","DOI":"10.1186\/s13244-023-01560-0","volume":"14","author":"DP dos Santos","year":"2023","unstructured":"dos Santos DP, Kotter E, Mildenberger P, Mart\u00ed-Bonmat\u00ed LEuropean Society of Radiology (ESR) (2023) ESR paper on structured reporting in radiology\u2014update 2023. Insights Imaging 14:199. https:\/\/doi.org\/10.1186\/s13244-023-01560-0.","journal-title":"Insights Imaging"},{"key":"804_CR18","doi-asserted-by":"publisher","first-page":"1406","DOI":"10.3174\/ajnr.A5697","volume":"39","author":"MD Mamlouk","year":"2018","unstructured":"Mamlouk MD, Chang PC, Saket RR (2018) Contextual radiology reporting: a new approach to neuroradiology structured templates. AJNR Am J Neuroradiol 39:1406\u20131414. https:\/\/doi.org\/10.3174\/ajnr.A5697","journal-title":"AJNR Am J Neuroradiol"},{"key":"804_CR19","doi-asserted-by":"publisher","unstructured":"Park J, Sim WS, Yu JY, Park YR, Lee YH (2025) Evaluation of context-aware prompting techniques for classification of tumor response categories in radiology reports using large language model. J Imaging Inform Med. https:\/\/doi.org\/10.1007\/s10278-025-01685-2","DOI":"10.1007\/s10278-025-01685-2"},{"key":"804_CR20","doi-asserted-by":"publisher","first-page":"1445","DOI":"10.1007\/s11604-025-01799-1","volume":"43","author":"H Takita","year":"2025","unstructured":"Takita H, Walston SL, Mitsuyama Y, Watanabe K, Ishimaru S, Ueda D (2025) Comparative performance of large language models in structuring head CT radiology reports: multi-institutional validation study in Japan. Jpn J Radiol 43:1445\u20131455. https:\/\/doi.org\/10.1007\/s11604-025-01799-1","journal-title":"Jpn J Radiol"},{"key":"804_CR21","doi-asserted-by":"publisher","unstructured":"Prucker P, Bressem KK, Kim SH et al (2025) Privacy-preserving generation of structured lymphoma progression reports from cross-sectional imaging: a comparative analysis of Llama 3.3 and Llama 4. J Imaging Inform Med. https:\/\/doi.org\/10.1007\/s10278-025-01618-z","DOI":"10.1007\/s10278-025-01618-z"},{"key":"804_CR22","doi-asserted-by":"publisher","first-page":"1120","DOI":"10.1093\/jamia\/ocaf056","volume":"32","author":"E Mahmoudi","year":"2025","unstructured":"Mahmoudi E, Vahdati S, Chao CJ et al (2025) A comparative analysis of privacy-preserving large language models for automated echocardiography report analysis. J Am Med Inform Assoc 32:1120\u20131129. https:\/\/doi.org\/10.1093\/jamia\/ocaf056","journal-title":"J Am Med Inform Assoc"},{"key":"804_CR23","doi-asserted-by":"publisher","first-page":"e67056","DOI":"10.2196\/67056","volume":"12","author":"S Kim","year":"2024","unstructured":"Kim S, Jang S, Kim B et al (2024) Automated pathologic TN classification prediction and rationale generation from lung cancer surgical pathology reports using a large language model fine-tuned with chain-of-thought: algorithm development and validation study. JMIR Med Inform 12:e67056. https:\/\/doi.org\/10.2196\/67056","journal-title":"JMIR Med Inform"},{"key":"804_CR24","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1007\/s00784-025-06498-9","volume":"29","author":"YT Xiong","year":"2025","unstructured":"Xiong YT, Lian WJ, Sun YN et al (2025) Exploring GPT-4o\u2019s multimodal reasoning capabilities with panoramic radiograph: the role of prompt engineering. Clin Oral Investig 29:405. https:\/\/doi.org\/10.1007\/s00784-025-06498-9","journal-title":"Clin Oral Investig"},{"key":"804_CR25","doi-asserted-by":"publisher","unstructured":"Wei J, Wang X, Schuurmans D et al (2022) Chain-of-thought prompting elicits reasoning in large language models. Adv Neural Inf Process 35:24824\u201324837 https:\/\/doi.org\/10.52202\/068431-1800","DOI":"10.52202\/068431-1800"},{"key":"804_CR26","doi-asserted-by":"publisher","first-page":"1602","DOI":"10.21037\/qims-23-1180","volume":"14","author":"Z Wang","year":"2024","unstructured":"Wang Z, Zhang Z, Traverso A, Dekker A, Qian L, Sun P (2024) Assessing the role of GPT-4 in thyroid ultrasound diagnosis and treatment recommendations: enhancing interpretability with a chain-of-thought approach. Quant Imaging Med Surg 14:1602\u20131615. https:\/\/doi.org\/10.21037\/qims-23-1180","journal-title":"Quant Imaging Med Surg"},{"key":"804_CR27","doi-asserted-by":"publisher","first-page":"783","DOI":"10.3233\/SHTI250466","volume":"327","author":"HA Shah","year":"2025","unstructured":"Shah HA, Househ M (2025) Chain of thought strategy for smaller LLMs for medical reasoning. Stud Health Technol Inform 327:783\u2013787. https:\/\/doi.org\/10.3233\/SHTI250466","journal-title":"Stud Health Technol Inform"},{"key":"804_CR28","doi-asserted-by":"publisher","first-page":"1929","DOI":"10.1093\/jamia\/ocae095","volume":"31","author":"M Li","year":"2024","unstructured":"Li M, Zhou H, Yang H, Zhang R (2024) RT: a retrieving and chain-of-thought framework for few-shot medical named entity recognition. J Am Med Inform Assoc 31:1929\u20131938. https:\/\/doi.org\/10.1093\/jamia\/ocae095","journal-title":"J Am Med Inform Assoc"},{"key":"804_CR29","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1038\/s44401-025-00018-3","volume":"2","author":"Z Hou","year":"2025","unstructured":"Hou Z, Liu H, Bian J, He X, Zhuang Y (2025) Enhancing medical coding efficiency through domain-specific fine-tuned large language models. NPJ Health Syst 2:14. https:\/\/doi.org\/10.1038\/s44401-025-00018-3","journal-title":"NPJ Health Syst"},{"key":"804_CR30","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.220119","volume":"4","author":"WF Wiggins","year":"2022","unstructured":"Wiggins WF, Tejani AS (2022) On the opportunities and risks of foundation models for natural language processing in radiology. Radiol Artif Intell 4:e220119. https:\/\/doi.org\/10.1148\/ryai.220119","journal-title":"Radiol Artif Intell"},{"key":"804_CR31","doi-asserted-by":"publisher","unstructured":"Mese I, Akinci D\u2019Antonoli T, Bluethgen C, et al (2026) Reporting checklist for foundation and large language models in medical research (REFINE): an international consensus guideline. Diagn Interv Radiol. https:\/\/doi.org\/10.4274\/dir.2026.263812","DOI":"10.4274\/dir.2026.263812"},{"key":"804_CR32","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1016\/j.jacr.2004.08.005","volume":"2","author":"DC Kushner","year":"2005","unstructured":"Kushner DC, Lucey LL (2005) Diagnostic radiology reporting and communication: the ACR Guideline. J Am Coll Radiol 2:15\u201321. https:\/\/doi.org\/10.1016\/j.jacr.2004.08.005","journal-title":"J Am Coll Radiol"},{"key":"804_CR33","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-022-01946-y","volume":"22","author":"J Li","year":"2022","unstructured":"Li J, Lin Y, Zhao P et al (2022) Automatic text classification of actionable radiology reports of tinnitus patients using bidirectional encoder representations from transformer (BERT) and in-domain pre-training (IDPT). BMC Med Inform Decis Mak 22:200. https:\/\/doi.org\/10.1186\/s12911-022-01946-y","journal-title":"BMC Med Inform Decis Mak"},{"key":"804_CR34","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1038\/s41586-023-06291-2","volume":"620","author":"K Singhal","year":"2023","unstructured":"Singhal K, Azizi S, Tu T et al (2023) Large language models encode clinical knowledge. Nature 620:172\u2013180. https:\/\/doi.org\/10.1038\/s41586-023-06291-2","journal-title":"Nature"},{"key":"804_CR35","doi-asserted-by":"publisher","first-page":"887","DOI":"10.3390\/healthcare11060887","volume":"11","author":"M Sallam","year":"2023","unstructured":"Sallam M (2023) ChatGPT utility in healthcare education, research, and practice: systematic review on the promising perspectives and valid concerns. Healthcare 11:887. https:\/\/doi.org\/10.3390\/healthcare11060887","journal-title":"Healthcare"},{"key":"804_CR36","doi-asserted-by":"publisher","unstructured":"Zhou D, Sch\u00e4rli N, Hou L, et al (2022) Least-to-most prompting enables complex reasoning in large language models. https:\/\/doi.org\/10.48550\/ARXIV.2205.10625","DOI":"10.48550\/ARXIV.2205.10625"},{"key":"804_CR37","doi-asserted-by":"publisher","first-page":"1158","DOI":"10.1016\/j.crad.2015.06.091","volume":"70","author":"NH Strickland","year":"2015","unstructured":"Strickland NH (2015) Quality assurance in radiology: peer review and peer feedback. Clin Radiol 70:1158\u20131164. https:\/\/doi.org\/10.1016\/j.crad.2015.06.091","journal-title":"Clin Radiol"},{"key":"804_CR38","doi-asserted-by":"crossref","unstructured":"Tabassi E (2023) Artificial intelligence risk management framework (AI RMF 1.0). National Institute of Standards and Technology. Available via https:\/\/www.nist.gov\/publications\/artificial-intelligence-risk-management-framework-ai-rmf-10. Accessed 4 Jan 2026","DOI":"10.6028\/NIST.AI.100-1"},{"key":"804_CR39","unstructured":"World Health Organization (2021) Ethics and governance of artificial intelligence for health. Available via https:\/\/www.who.int\/publications\/i\/item\/9789240029200. Accessed 4 Jan 2026"},{"key":"804_CR40","doi-asserted-by":"publisher","first-page":"1080","DOI":"10.1016\/j.jacr.2017.03.023","volume":"14","author":"S Goldberg-Stein","year":"2017","unstructured":"Goldberg-Stein S, Frigini LA, Long S et al (2017) ACR RADPEER Committee White Paper with 2016 updates: revised scoring system, new classifications, self-review, and subspecialized reports. J Am Coll Radiol 14:1080\u20131086. https:\/\/doi.org\/10.1016\/j.jacr.2017.03.023","journal-title":"J Am Coll Radiol"},{"key":"804_CR41","doi-asserted-by":"publisher","unstructured":"Lanham T, Chen A, Radhakrishnan A et al (2023) Measuring faithfulness in chain-of-thought reasoning. https:\/\/doi.org\/10.48550\/ARXIV.2307.13702","DOI":"10.48550\/ARXIV.2307.13702"},{"key":"804_CR42","doi-asserted-by":"publisher","first-page":"74952","DOI":"10.52202\/075280-3275","volume":"36","author":"M Turpin","year":"2023","unstructured":"Turpin M, Michael J, Perez E, Bowman S (2023) Language models don\u2019t always say what they think: unfaithful explanations in chain-of-thought prompting. Adv Neural Inf Process Syst 36:74952\u201374965","journal-title":"Adv Neural Inf Process Syst"}],"container-title":["European Radiology Experimental"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s41747-026-00804-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s41747-026-00804-0","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s41747-026-00804-0.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,22]],"date-time":"2026-09-22T10:11:46Z","timestamp":1790071906000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s41747-026-00804-0"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,22]]},"references-count":42,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["804"],"URL":"https:\/\/doi.org\/10.1186\/s41747-026-00804-0","relation":{},"ISSN":["2509-9280"],"issn-type":[{"value":"2509-9280","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,22]]},"assertion":[{"value":"26 January 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"28 June 2026","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"19 July 2026","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 September 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The institutional review board of Beijing Friendship Hospital, Capital Medical University, waived written informed consent (approval\/waiver ID: 2021-P2-142-01) because all reports were de-identified and the study posed minimal risk.","order":1,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"All authors read the manuscript and agreed to publish.","order":2,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no conflict of interest.","order":3,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"141"}}