{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,23]],"date-time":"2026-10-23T19:14:51Z","timestamp":1792782891279,"version":"4.5.0"},"reference-count":26,"publisher":"Springer Science and Business Media LLC","license":[{"start":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T00:00:00Z","timestamp":1790121600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T00:00:00Z","timestamp":1790121600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Loma Linda University","id":[{"id":"https:\/\/ror.org\/04bj28v14","id-type":"ROR","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["J Digit Imaging. Inform. med."],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Large language models (LLMs) answer medical questions fluently but fabricate statements and citations too often to be trusted unaided at the point of care. We describe RADAR (Radiology Appropriateness Decision and Advisory Resource), a prototype retrieval-augmented consult-support pipeline that grounds an LLM in a knowledge base structured like the American College of Radiology (ACR) Appropriateness Criteria at the level of individual clinical variants. The knowledge base is synthetic (twelve variants across five topics) because the official criteria are copyrighted. The design is deliberately conservative: a BM25 retriever, an absolute-score abstention gate, a structured-output generator bound by an in-prompt grounding contract, and a deterministic validator that strips any recommendation whose citation does not resolve to a retrieved chunk. We report a development set (eight scenarios, seven off-domain probes) used during parameter selection and a held-out set written after all parameters were frozen (36 scenarios, three per variant, and 24 clinically adjacent probes), evaluated in a deterministic retrieval-only mode and in the full pipeline over five independent runs at temperature 0. Retrieval alone ranked the correct variant first in 25 of 36 held-out scenarios (69.4%; 95% CI 51.9\u201383.7) but always placed it within the top five, and the score gate stopped only 3 of 22 clinically adjacent probes. The full pipeline (claude-sonnet-4\u20136, temperature 0, five runs) selected the correct variant and top procedure in every run for 33 of 36 scenarios (91.7%; 95% CI 77.5\u201398.2) and in 33 or 34 of 36 scenarios in each individual run; every failed run was a false abstention on a thunderclap-headache presentation, and no run of any scenario recommended a wrong variant. It refused 17 of 22 clinically adjacent probes (77.3%; 54.6\u201392.2). All 721 emitted citations resolved to a retrieved chunk. The findings characterize a proof of concept: lexical retrieval delivers topic routing and citation traceability, while variant selection and realistic abstention depend on the generator acting as a semantic reranker, and the residual failures are refusals rather than wrong answers. Clinical readiness will require licensed ACR content, a larger reference-standard set, and prospective evaluation.<\/jats:p>","DOI":"10.1007\/s10278-026-02326-y","type":"journal-article","created":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T18:26:27Z","timestamp":1790187987000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["RADAR: A Prototype for Grounding Large Language Model Consult Support in an ACR-Structured Appropriateness Knowledge Base with Citation Validation and Abstention"],"prefix":"10.1007","author":[{"given":"Chiranjeevi","family":"Vantaku","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-1387-9703","authenticated-orcid":false,"given":"Amy","family":"Avakian","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,9,23]]},"reference":[{"key":"2326_CR1","doi-asserted-by":"publisher","first-page":"S240","DOI":"10.1016\/j.jacr.2021.03.021","volume":"18","author":"DA Kurth","year":"2021","unstructured":"Kurth DA, Karmazyn BK, Waldrip CA, Chatfield M, Lockhart ME: ACR Appropriateness Criteria\u00ae methodology. J Am Coll Radiol 18:S240-S250, 2021","journal-title":"J Am Coll Radiol"},{"key":"2326_CR2","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1016\/j.jacr.2015.08.026","volume":"13","author":"AY Sheng","year":"2016","unstructured":"Sheng AY, Castro A, Lewiss RE: Awareness, utilization, and education of the ACR Appropriateness Criteria: a review and future directions. J Am Coll Radiol 13:131-136, 2016","journal-title":"J Am Coll Radiol"},{"key":"2326_CR3","doi-asserted-by":"publisher","first-page":"147","DOI":"10.1148\/radiol.2511081174","volume":"251","author":"CL Sistrom","year":"2009","unstructured":"Sistrom CL, Dang PA, Weilburg JB, Dreyer KJ, Rosenthal DI, Thrall JH: Effect of computerized order entry with integrated decision support on the growth of outpatient procedure volumes: seven-year time series analysis. Radiology 251:147-155, 2009","journal-title":"Radiology"},{"key":"2326_CR4","doi-asserted-by":"publisher","first-page":"799","DOI":"10.1016\/j.jacr.2006.05.006","volume":"3","author":"DI Rosenthal","year":"2006","unstructured":"Rosenthal DI, Weilburg JB, Schultz T, et al: Radiology order entry with decision support: initial clinical experience. J Am Coll Radiol 3:799-806, 2006","journal-title":"J Am Coll Radiol"},{"key":"2326_CR5","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: Large language models encode clinical knowledge. Nature 620:172-180, 2023","journal-title":"Nature"},{"key":"2326_CR6","doi-asserted-by":"publisher","first-page":"943","DOI":"10.1038\/s41591-024-03423-7","volume":"31","author":"K Singhal","year":"2025","unstructured":"Singhal K, Tu T, Gottweis J, et al: Toward expert-level medical question answering with large language models. Nat Med 31:943-950, 2025","journal-title":"Nat Med"},{"key":"2326_CR7","unstructured":"Nori H, King N, McKinney SM, Carignan D, Horvitz E: Capabilities of GPT-4 on medical challenge problems.\u00a0arXiv:2303.13375, 2023"},{"key":"2326_CR8","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.230582","volume":"307","author":"R Bhayana","year":"2023","unstructured":"Bhayana R, Krishna S, Bleakney RR: Performance of ChatGPT on a radiology board-style examination: insights into current strengths and limitations. Radiology 307:e230582, 2023","journal-title":"Radiology"},{"key":"2326_CR9","doi-asserted-by":"publisher","DOI":"10.1016\/j.clinimag.2025.110573","volume":"125","author":"E \u00c7amur","year":"2025","unstructured":"\u00c7amur E, Cesur T, G\u00fcne\u015f YC: Evaluating large language models for imaging modality selection: potential to reduce unnecessary contrast agent use and radiation exposure. Clin Imaging 125:110573, 2025","journal-title":"Clin Imaging"},{"key":"2326_CR10","doi-asserted-by":"publisher","first-page":"299","DOI":"10.1016\/j.jacr.2025.11.023","volume":"23","author":"Y Yasar","year":"2026","unstructured":"Yasar Y, Demir M, Turgan AH, Cant\u00fcrk A: GPT-4o in interventional radiology decision-making: a simulation study using ACR Appropriateness Criteria. J Am Coll Radiol 23:299-302, 2026","journal-title":"J Am Coll Radiol"},{"issue":"12","key":"2326_CR11","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3571730","volume":"55","author":"Z Ji","year":"2023","unstructured":"Ji Z, Lee N, Frieske R, et al: Survey of hallucination in natural language generation. ACM Comput Surv 55(12):1-38, 2023","journal-title":"ACM Comput Surv"},{"key":"2326_CR12","doi-asserted-by":"publisher","first-page":"777","DOI":"10.1162\/coli_a_00486","volume":"49","author":"H Rashkin","year":"2023","unstructured":"Rashkin H, Nikolaev V, Lamm M, et al: Measuring attribution in natural language generation models. Comput Linguist 49:777-840, 2023","journal-title":"Comput Linguist"},{"key":"2326_CR13","unstructured":"Bohnet B, Tran VQ, Verga P, et al: Attributed question answering: evaluation and modeling for attributed large language models.\u00a0arXiv:2212.08037, 2022"},{"key":"2326_CR14","first-page":"173","volume":"32","author":"YC G\u00fcne\u015f","year":"2026","unstructured":"G\u00fcne\u015f YC, Cesur T, \u00c7amur E: Evaluating the reference accuracy of large language models in radiology: a comparative study across subspecialties. Diagn Interv Radiol 32:173-181, 2026","journal-title":"Diagn Interv Radiol"},{"key":"2326_CR15","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.40969","volume":"7","author":"E Goh","year":"2024","unstructured":"Goh E, Gallo R, Hom J, et al: Large language model influence on diagnostic reasoning: a randomized clinical trial. JAMA Netw Open 7:e2440969, 2024","journal-title":"JAMA Netw Open"},{"key":"2326_CR16","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.230970","volume":"308","author":"A Rau","year":"2023","unstructured":"Rau A, Rau S, Zoeller D, et al: A context-based chatbot surpasses radiologists and generic ChatGPT in following the ACR Appropriateness Guidelines. Radiology 308:e230970, 2023","journal-title":"Radiology"},{"key":"2326_CR17","first-page":"9459","volume":"33","author":"P Lewis","year":"2020","unstructured":"Lewis P, Perez E, Piktus A, et al: Retrieval-augmented generation for knowledge-intensive NLP tasks. Adv Neural Inf Process Syst 33:9459-9474, 2020","journal-title":"Adv Neural Inf Process Syst"},{"key":"2326_CR18","unstructured":"Guu K, Lee K, Tung Z, Pasupat P, Chang MW: REALM: retrieval-augmented language model pre-training. Proc Int Conf Mach Learn (ICML), 2020"},{"key":"2326_CR19","unstructured":"Borgeaud S, Mensch A, Hoffmann J, et al: Improving language models by retrieving from trillions of tokens. Proc Int Conf Mach Learn (ICML), 2022"},{"key":"2326_CR20","doi-asserted-by":"crossref","unstructured":"Gao L, Dai Z, Pasupat P, et al: RARR: researching and revising what language models say, using language models. Proc 61st Annu Meeting Assoc Comput Linguist (ACL):16477\u201316508, 2023","DOI":"10.18653\/v1\/2023.acl-long.910"},{"key":"2326_CR21","doi-asserted-by":"publisher","first-page":"529","DOI":"10.1162\/tacl_a_00754","volume":"13","author":"B Wen","year":"2025","unstructured":"Wen B, Yao J, Feng S, et al: Know your limits: a survey of abstention in large language models. Trans Assoc Comput Linguist 13:529-556, 2025","journal-title":"Trans Assoc Comput Linguist"},{"key":"2326_CR22","first-page":"2024","volume":"6233\u20136251","author":"G Xiong","year":"2024","unstructured":"Xiong G, Jin Q, Lu Z, Zhang A: Benchmarking retrieval-augmented generation for medicine. Findings Assoc Comput Linguist: ACL 2024:6233-6251, 2024","journal-title":"Findings Assoc Comput Linguist: ACL"},{"key":"2326_CR23","doi-asserted-by":"publisher","first-page":"157","DOI":"10.1162\/tacl_a_00638","volume":"12","author":"NF Liu","year":"2024","unstructured":"Liu NF, Lin K, Hewitt J, et al: Lost in the middle: how language models use long contexts. Trans Assoc Comput Linguist 12:157-173, 2024","journal-title":"Trans Assoc Comput Linguist"},{"key":"2326_CR24","doi-asserted-by":"crossref","unstructured":"Zakka C, Shad R, Chaurasia A, et al: Almanac\u2014retrieval-augmented language models for clinical medicine. NEJM AI 1(2), 2024","DOI":"10.1056\/AIoa2300068"},{"key":"2326_CR25","first-page":"333","volume":"3","author":"S Robertson","year":"2009","unstructured":"Robertson S, Zaragoza H: The probabilistic relevance framework: BM25 and beyond. Found Trends Inf Retr 3:333-389, 2009","journal-title":"Found Trends Inf Retr"},{"key":"2326_CR26","doi-asserted-by":"publisher","DOI":"10.1016\/j.acra.2026.01.017","author":"Y Yasar","year":"2026","unstructured":"Yasar Y, Demir M, Canturk A, et al: Effect of ChatGPT-assisted reflective reasoning on guideline-concordant procedural decision-making among early-career interventional radiologists. Acad Radiol, 2026. https:\/\/doi.org\/10.1016\/j.acra.2026.01.017","journal-title":"Acad Radiol"}],"container-title":["Journal of Imaging Informatics in Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02326-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s10278-026-02326-y","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s10278-026-02326-y.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T18:26:29Z","timestamp":1790187989000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s10278-026-02326-y"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,23]]},"references-count":26,"alternative-id":["2326"],"URL":"https:\/\/doi.org\/10.1007\/s10278-026-02326-y","relation":{},"ISSN":["2948-2933"],"issn-type":[{"value":"2948-2933","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,23]]},"assertion":[{"value":"23 July 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"11 September 2026","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"14 September 2026","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"23 September 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"This study did not involve human participants, human data, human biological material, or animals. The knowledge base is synthetic and all consult scenarios are fictional constructs written for evaluation purposes; no patient data of any kind was used. Ethics approval was therefore not required.","order":1,"name":"Ethics","label":"Ethics Approval","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"Not applicable; no human participants were involved.","order":2,"name":"Ethics","label":"Consent to Participate\/Consent for Publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}]}}