{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,15]],"date-time":"2026-03-15T09:56:21Z","timestamp":1773568581047,"version":"3.50.1"},"reference-count":7,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2024,10,3]],"date-time":"2024-10-03T00:00:00Z","timestamp":1727913600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2024,10,3]],"date-time":"2024-10-03T00:00:00Z","timestamp":1727913600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"},{"start":{"date-parts":[[2024,10,3]],"date-time":"2024-10-03T00:00:00Z","timestamp":1727913600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2024,10,3]],"date-time":"2024-10-03T00:00:00Z","timestamp":1727913600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"}],"funder":[{"DOI":"10.13039\/100008894","name":"Stryker","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100008894","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["clinicalkey.fr","clinicalkey.jp","arthroscopyjournal.org","clinicalkey.es","clinicalkey.com.au","clinicalkey.com","elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Arthroscopy"],"published-print":{"date-parts":[[2025,2]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>The content accuracy of off\u2010the\u2010shelf large language models (LLMs) mirrors the content accuracy of the unregulated Internet from which these generative artificial intelligence models are supplied. With error rates approximating 30% in terms of treatment recommendations for the management of common musculoskeletal conditions, seeking expert opinion remains paramount. However, custom LLMs represent an excellent opportunity to infuse niche, bespoke expertise from the many specialties and subspecialties within medicine. Methods of customizing these generative models broadly fall under the categories of prompt engineering; \u201cretrieval\u2010augmented generation\u201d prioritizing retrieval of relevant information from a specific domain of data; \u201cfine\u2010tuning\u201d of a basic pretrained model into one that is refined for health care\u2013related vernacular and acronyms; and \u201cagentic augmentation\u201d including software that breaks down complex tasks into smaller ones, recruiting multiple LLMs (with or without retrieval\u2010augmented generation), optimizing the output, internally deciding whether the response is appropriate or sufficient, and even passing on an unmet outcome to a human for supervision (\u201cphone a friend\u201d). Custom LLMs offer physicians and their associated organizations the rare opportunity to regain control of our profession by re\u2010establishing authority in our increasingly digital landscape.<\/jats:p>","DOI":"10.1016\/j.arthro.2024.09.047","type":"journal-article","created":{"date-parts":[[2024,10,3]],"date-time":"2024-10-03T11:50:48Z","timestamp":1727956248000},"page":"276-278","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":11,"title":["<i>Editorial Commentary:<\/i>\n                    Off\u2010the\u2010Shelf Large Language Models Are of Insufficient Quality to Provide Medical Treatment Recommendations, While Customization of Large Language Models Results in Quality Recommendations"],"prefix":"10.1002","volume":"41","author":[{"given":"Prem N.","family":"Ramkumar","sequence":"first","affiliation":[{"name":"Commons Clinic"},{"name":"The Warren Alpert Medical School of Brown University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Andrew F.","family":"Masotto","sequence":"additional","affiliation":[{"name":"Commons Clinic"},{"name":"The Warren Alpert Medical School of Brown University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Joshua J.","family":"Woo","sequence":"additional","affiliation":[{"name":"Commons Clinic"},{"name":"The Warren Alpert Medical School of Brown University"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"311","published-online":{"date-parts":[[2024,10,3]]},"reference":[{"key":"e_1_2_3_2_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.arthro.2024.07.040"},{"key":"e_1_2_3_3_2","doi-asserted-by":"publisher","DOI":"10.1186\/s40634-023-00700-1"},{"key":"e_1_2_3_4_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.arth.2023.04.007"},{"key":"e_1_2_3_5_2","first-page":"2983","article-title":"Large language models should be used as scientific reasoning engines, not knowledge databases","volume":"29","author":"Truhn D.","year":"2023","journal-title":"Nature"},{"key":"e_1_2_3_6_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06291-2"},{"key":"e_1_2_3_7_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-024-01239-w"},{"key":"e_1_2_3_8_2","doi-asserted-by":"crossref","unstructured":"Woo JJ Yang AJ Olsen RJ et\u00a0al. Custom large language models improve accuracy: Comparing retrieval augmented generation and artificial intelligence agents to non\u2010custom models for evidence\u2010based medicine. [published online November 7 2024]Arthroscopy. doi:10.1016\/j.arthro.2024.10.042","DOI":"10.1016\/j.arthro.2024.10.042"}],"container-title":["Arthroscopy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0749806324007709?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0749806324007709?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/arthroscopyjournals.onlinelibrary.wiley.com\/doi\/pdf\/10.1016\/j.arthro.2024.09.047","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/arthroscopyjournals.onlinelibrary.wiley.com\/doi\/full-xml\/10.1016\/j.arthro.2024.09.047","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/arthroscopyjournals.onlinelibrary.wiley.com\/doi\/pdf\/10.1016\/j.arthro.2024.09.047","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,15]],"date-time":"2026-03-15T08:52:15Z","timestamp":1773564735000},"score":1,"resource":{"primary":{"URL":"https:\/\/arthroscopyjournals.onlinelibrary.wiley.com\/doi\/10.1016\/j.arthro.2024.09.047"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,3]]},"references-count":7,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,2]]}},"alternative-id":["10.1016\/j.arthro.2024.09.047"],"URL":"https:\/\/doi.org\/10.1016\/j.arthro.2024.09.047","archive":["Portico"],"relation":{},"ISSN":["0749-8063","1526-3231"],"issn-type":[{"value":"0749-8063","type":"print"},{"value":"1526-3231","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,10,3]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"Editorial Commentary: Off-the-Shelf Large Language Models Are of Insufficient Quality to Provide Medical Treatment Recommendations, While Customization of Large Language Models Results in Quality Recommendations","name":"articletitle","label":"Article Title"},{"value":"Arthroscopy: The Journal of Arthroscopic & Related Surgery","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.arthro.2024.09.047","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"https:\/\/doi.org\/10.1016\/j.arthro.2024.07.040","name":"associatedlink","label":"CrossRef DOI link to the associated document"},{"value":"simple-article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2024 by the Arthroscopy Association of North America","name":"copyright","label":"Copyright"}]}}