{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,2]],"date-time":"2026-05-02T13:56:33Z","timestamp":1777730193003,"version":"3.51.4"},"reference-count":19,"publisher":"SAGE Publications","license":[{"start":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T00:00:00Z","timestamp":1764028800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["Interv Neuroradiol"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>As large language models (LLMs) become increasingly accessible to the public, patients are turning to these tools for medical guidance \u2014 including in highly specialized fields like interventional neuroradiology. Despite their growing use, the safety, completeness, and reliability of LLM-generated information in subspecialty medicine remain unclear.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Five publicly available LLMs \u2014 ChatGPT, Gemini, Claude, Perplexity, and DeepSeek \u2014 were prompted with four neurointerventional patient-facing clinical questions spanning ischemic stroke, hemorrhagic stroke, venous disorders, and procedural device use. Each model was queried three times per question to generate unique responses. Eight blinded raters scored each response on accuracy, completeness, safety, and actionability using Likert scales. Plagiarism analyses were also performed.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>DeepSeek consistently outperformed other LLMs in accuracy, completeness, and actionability across four prompts, while Gemini frequently ranked worse, including in plagiarism levels. ChatGPT performed well in accuracy. Physicians were more critical than non-physicians across accuracy, completeness, and safety, whereas non-physicians rated actionability significantly lower. Overall, LLMs were rated relatively high (median of &gt;4 on a 5-point scale) in medical safety, suggesting low risk of overtly harmful advice.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Recent-generation LLMs offer medically safe, though often incomplete or imprecise, information in response to patient-oriented neurointerventional queries. Including non-physician raters revealed valuable differences in perception that are relevant to how patients may interpret LLM outputs. As benchmark frameworks like HealthBench improve LLM evaluation, inclusion of lay perspectives and subspecialty contexts remains essential. Responsible use by clinicians and ongoing patient education will be critical as LLM use in healthcare expands.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1177\/15910199251396358","type":"journal-article","created":{"date-parts":[[2025,11,25]],"date-time":"2025-11-25T18:11:49Z","timestamp":1764094309000},"update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":2,"title":["Large language model responses to patient-oriented neurointerventional queries: A multirater assessment of accuracy, completeness, safety, and actionability"],"prefix":"10.1177","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3284-7284","authenticated-orcid":false,"given":"Albert HW","family":"Jiang","sequence":"first","affiliation":[{"name":"University of Hawai\u02bbi at M\u0101noa"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7047-9676","authenticated-orcid":false,"given":"Tyler R","family":"Ray","sequence":"additional","affiliation":[{"name":"University of Hawai\u02bbi at M\u0101noa"},{"name":"Cell and Molecular Biology, John A. Burns School of Medicine, University of Hawai\u02bbi at M\u0101noa, Honolulu, HI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-9807-8990","authenticated-orcid":false,"given":"Alisha E","family":"Suri","sequence":"additional","affiliation":[{"name":"University of Hawai\u02bbi at M\u0101noa"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andrew R","family":"Menard","sequence":"additional","affiliation":[{"name":"Johns Hopkins HealthCare LLC"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6491-5443","authenticated-orcid":false,"given":"Ryan T","family":"Kellogg","sequence":"additional","affiliation":[{"name":"University of Virginia"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6172-2010","authenticated-orcid":false,"given":"Arindam R","family":"Chatterjee","sequence":"additional","affiliation":[{"name":"Washington University School of Medicine"},{"name":"Washington University School of Medicine, St Louis, MO, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Matthew A","family":"Koenig","sequence":"additional","affiliation":[{"name":"University of Hawai\u02bbi at M\u0101noa"},{"name":"Neurology, The Queen's Health Systems"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-5405-1966","authenticated-orcid":false,"given":"Roy K","family":"Esaki","sequence":"additional","affiliation":[{"name":"Anesthesiology, The Queen\u2019s Health Systems, Honolulu, HI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3759-7886","authenticated-orcid":false,"given":"Ferdinand K","family":"Hui","sequence":"additional","affiliation":[{"name":"Neurointerventional Surgery, The Queen\u2019s Health Systems, Honolulu, HI, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7164-1479","authenticated-orcid":false,"given":"Jan","family":"Vargas","sequence":"additional","affiliation":[{"name":"Prima Health Upstate"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"179","published-online":{"date-parts":[[2025,11,25]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0109583"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2011.12.004"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1038\/s43856-024-00717-2"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.36922\/aih.4808"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.0347"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1001\/jamainternmed.2023.1838"},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","DOI":"10.1111\/imj.16607"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10916-023-01925-4"},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01074-z"},{"key":"e_1_3_2_11_2","doi-asserted-by":"publisher","DOI":"10.2196\/70315"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2023.5321"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1177\/15910199221122848"},{"key":"e_1_3_2_14_2","doi-asserted-by":"publisher","DOI":"10.1177\/15910199221150470"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.1177\/15910199221097475"},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01029-4"},{"key":"e_1_3_2_17_2","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2024.1495582"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-024-02656-3"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.2196\/60083"},{"key":"e_1_3_2_20_2","unstructured":"Arora RK Wei J Hicks RS et al. HealthBench: Evaluating Large Language Models Towards Improved Human Health. OpenAI https:\/\/cdn.openai.com\/pdf\/bd7a39d5-9e9f-47b3-903c-8b847ca650c7\/healthbench_paper.pdf (2025)."}],"updated-by":[{"DOI":"10.1177\/15910199261444853","type":"expression_of_concern","label":"Expression of concern","source":"publisher","updated":{"date-parts":[[2026,4,25]],"date-time":"2026-04-25T00:00:00Z","timestamp":1777075200000}}],"container-title":["Interventional Neuroradiology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/15910199251396358","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/15910199251396358","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/15910199251396358","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T10:57:44Z","timestamp":1777460264000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/15910199251396358"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,25]]},"references-count":19,"alternative-id":["10.1177\/15910199251396358"],"URL":"https:\/\/doi.org\/10.1177\/15910199251396358","relation":{},"ISSN":["1591-0199","2385-2011"],"issn-type":[{"value":"1591-0199","type":"print"},{"value":"2385-2011","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,11,25]]},"article-number":"15910199251396358"}}