{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,7]],"date-time":"2026-09-07T14:35:32Z","timestamp":1788791732749,"version":"build-2803163510"},"reference-count":26,"publisher":"SAGE Publications","issue":"6","license":[{"start":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T00:00:00Z","timestamp":1766361600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"},{"start":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T00:00:00Z","timestamp":1766361600000},"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":["Global Spine Journal"],"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:sec>\n                    <jats:title>Study Design<\/jats:title>\n                    <jats:p>Vignette-based cross-sectional study.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>Generative artificial intelligence (AI) programs such as large language models (LLMs) are reshaping treatment decision-making, yet applications in minimally invasive spine surgery (MISS) are still scarce. This study examines whether OpenAI\u2019s ChatGPT-5 Pro and Google\u2019s Gemini 2.5 Pro reproduce expert management categories from published MISS cases and measures agreement at procedural and binary triage levels.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We constructed 90 clinical vignettes from published case reports and prompted each LLM to assign 1 or more of ten predefined categories with two-sentence rationales. Agreement with reference was assessed using Jensen\u2013Shannon divergence (JSD), Stuart\u2013Maxwell tests, Cohen\u2019s \u03ba, and McNemar\u2019s test for surgical vs non-surgical triage.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Divergence from reference was small, with Jensen\u2013Shannon divergence 0.115 (ChatGPT-5 Pro) and 0.112 (Gemini 2.5 Pro), and smaller between models at 0.073. Paired multinomial tests found differences from the reference (Stuart\u2013Maxwell \u03c7\n                      <jats:sup>2<\/jats:sup>\n                      (9) = 24.8 and 26.0;\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      = 0.007 and 0.006) but not between models (14.4;\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      = 0.108). Case-level agreement was slight for ChatGPT-5 Pro and fair for Gemini 2.5 Pro (\u03ba = 0.146 and 0.245). Collapsing categories to surgical vs non-surgical improved agreement (\u03ba = 0.415 and 0.587 vs reference; 0.692 between models) with no bias in rates (\n                      <jats:italic toggle=\"yes\">P<\/jats:italic>\n                      \u2265 0.401).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>LLMs may differentiate between surgical and non-surgical triage, but procedure selection should remain expert-led until systems mature. These findings establish a baseline for integrating LLMs into surgical triage workflows and highlight promise and limitations of generative AI in precision spine care.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1177\/21925682251411225","type":"journal-article","created":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T09:14:28Z","timestamp":1766394868000},"page":"2611-2621","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":4,"title":["Evaluating Large Language Models for Decision Support in Minimally Invasive Spine Surgery Triage and Procedural Categories"],"prefix":"10.1177","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-4095-0798","authenticated-orcid":false,"given":"Ahmet","family":"Kartal","sequence":"first","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Noel F.","family":"Manalil","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chiungwen D.","family":"Cheng","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lawrance K.","family":"Chung","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Harry","family":"Gebhard","sequence":"additional","affiliation":[{"name":"Medical University Lausitz \u2013 Carl Thiem"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Greenberg","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roger","family":"H\u00e4rtl","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Galal A.","family":"Elsayed","sequence":"additional","affiliation":[{"name":"Och Spine at NewYork-Presbyterian Hospital"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2025,12,22]]},"reference":[{"key":"e_1_3_7_2_2","doi-asserted-by":"publisher","DOI":"10.14245\/ns.2347052.526"},{"key":"e_1_3_7_3_2","doi-asserted-by":"publisher","DOI":"10.1177\/21925682231195783"},{"key":"e_1_3_7_4_2","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.71105"},{"key":"e_1_3_7_5_2","doi-asserted-by":"publisher","DOI":"10.1097\/ms9.0000000000003519"},{"key":"e_1_3_7_6_2","doi-asserted-by":"publisher","DOI":"10.1007\/s00586-025-08825-w"},{"key":"e_1_3_7_7_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01684-1"},{"key":"e_1_3_7_8_2","doi-asserted-by":"publisher","DOI":"10.3389\/fsurg.2022.1054031"},{"key":"e_1_3_7_9_2","doi-asserted-by":"publisher","DOI":"10.1093\/neuros\/nyz464"},{"key":"e_1_3_7_10_2","doi-asserted-by":"publisher","DOI":"10.3171\/2014.3.Focus1413"},{"key":"e_1_3_7_11_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.spinee.2020.06.001"},{"key":"e_1_3_7_12_2","doi-asserted-by":"publisher","DOI":"10.3171\/2019.7.Spine181104"},{"key":"e_1_3_7_13_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01651-w"},{"key":"e_1_3_7_14_2","doi-asserted-by":"publisher","DOI":"10.1177\/20552076251337177"},{"key":"e_1_3_7_15_2","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocaf008"},{"key":"e_1_3_7_16_2","doi-asserted-by":"publisher","DOI":"10.1348\/000711006x126600"},{"key":"e_1_3_7_17_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.mex.2023.102212"},{"key":"e_1_3_7_18_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2003.813506"},{"key":"e_1_3_7_19_2","first-page":"2633","volume-title":"Extracting Training Data From Large Language Models","author":"Carlini N","year":"2021","unstructured":"Carlini N, Tramer F, Wallace E, et al. Extracting Training Data From Large Language Models; 2021:2633-2650."},{"key":"e_1_3_7_20_2","doi-asserted-by":"crossref","unstructured":"Dong Y Jiang X Liu H et al. Generalization or memorization: data contamination and trustworthy evaluation for large language models. arXiv preprint arXiv:240215938. 2024.","DOI":"10.18653\/v1\/2024.findings-acl.716"},{"key":"e_1_3_7_21_2","doi-asserted-by":"crossref","unstructured":"Fu Y Uzuner O Yetisgen M Xia F. Does data contamination detection work (well) for llms? A survey and evaluation on detection assumptions. arXiv preprint arXiv:241018966. 2024.","DOI":"10.18653\/v1\/2025.findings-naacl.291"},{"key":"e_1_3_7_22_2","unstructured":"Cloud G. Generative AI on Vertex AI: video understanding. https:\/\/cloud.google.com\/vertex-ai\/generative-ai\/docs\/models\/gemini\/2-5-pro. Accessed 15 September 2025."},{"key":"e_1_3_7_23_2","unstructured":"Developers GAf. Video understanding | Gemini API. https:\/\/ai.google.dev\/gemini-api\/docs\/video-understanding. Accessed 15 September 2025."},{"key":"e_1_3_7_24_2","unstructured":"Baddepudi A Yang A Lu\u010di\u0107 M. Advancing the frontier of video understanding with Gemini 2.5. https:\/\/developers.googleblog.com\/en\/gemini-2-5-video-understanding. Accessed 15 September 2025."},{"key":"e_1_3_7_25_2","unstructured":"OpenAI. Hello GPT-4o. https:\/\/openai.com\/index\/hello-gpt-4o. Accessed September 15 2025."},{"key":"e_1_3_7_26_2","unstructured":"OpenAI. Introducing gpt-realtime and Realtime API updates for production voice agents. https:\/\/openai.com\/index\/introducing-gpt-realtime. Accessed 15 September 2025."},{"key":"e_1_3_7_27_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-024-03097-1"}],"container-title":["Global Spine Journal"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/21925682251411225","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/21925682251411225","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/21925682251411225","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,11]],"date-time":"2026-07-11T07:49:54Z","timestamp":1783756194000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/21925682251411225"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,22]]},"references-count":26,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1177\/21925682251411225"],"URL":"https:\/\/doi.org\/10.1177\/21925682251411225","relation":{},"ISSN":["2192-5682","2192-5690"],"issn-type":[{"value":"2192-5682","type":"print"},{"value":"2192-5690","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,22]]}}}