{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T05:28:55Z","timestamp":1786339735082,"version":"build-2736575974"},"reference-count":28,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Vestibular disorders are common, burdensome, and frequently misdiagnosed, particularly in nonspecialist settings where history-taking is often incomplete or inconsistently structured. Digital health tools that standardize symptom elicitation could improve diagnostic triage, but most existing systems rely on static questionnaires or rule-based logic. Large language models (LLMs) offer a more flexible alternative through adaptive, natural-language consultations, but prospective evidence from real clinical workflows remains scarce.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to benchmark LLM diagnostic performance using static vestibular histories and to prospectively evaluate a locally deployed conversational LLM agent embedded in outpatient vertigo clinics.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We conducted a 2-phase diagnostic accuracy study. In the history-based evaluation (HBE), 10 LLMs and 5 senior otolaryngologists independently reviewed 227 structured vertigo histories, including 138 real patient questionnaires and 89 expert-simulated cases. In the prospective clinical evaluation (PCE), 176 outpatients with vertigo or dizziness were included in the analytic cohort across 5 centers in China. A nurse-assisted, tablet-based agent powered by DeepSeek-R1, deployed locally within institutional infrastructure, conducted multiturn symptom-history dialogues. The agent received history information only and did not receive physical examination or ancillary test findings. Attending clinicians were blinded to the agent\u2019s output and recorded final clinical diagnoses after routine assessment. The primary outcome was Top-1 diagnostic concordance with the reference diagnosis; secondary outcomes included HBE Top-3 accuracy and disorder-specific performance.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n \n In the HBE, Gemini-2.5-pro and o1 achieved the highest Top-1 accuracy (69.6%, 95% CI 63.2%\u201075.5% for each), and no LLM significantly outperformed the specialist panel majority vote (63.9%, 95% CI 57.3%\u201070.1%; McNemar test,\n P<\/jats:italic>\n >.10 for all models). In the PCE, the median age was 51 (IQR 38\u201061) years, and 128 of 176 (72.7%) participants were women. The conversational agent matched the reference diagnosis in 140 of 176 (79.55%; 95% CI 72.8%\u201085.2%) patients. Concordance was highest for benign paroxysmal positional vertigo (43\/44, 97.73%; 95% CI 88.0%\u201099.9%) and M\u00e9ni\u00e8re disease (27\/30, 90.00%; 95% CI 73.5%\u201097.9%) and lower for vestibular migraine (32\/45, 71.11%; 95% CI 55.7%\u201083.6%) and persistent postural-perceptual dizziness (19\/28, 67.86%; 95% CI 47.6%\u201084.1%).\n <\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n A locally deployed, history-only conversational LLM agent, operated within a nurse-assisted outpatient workflow, achieved approximately 80% concordance with specialist final diagnoses in a prospective multicenter study, with particularly high performance for benign paroxysmal positional vertigo. The concordance reflects the performance of the nurse-assisted agent workflow rather than the agent in isolation. These findings support the development of conversational LLMs as clinician-facing tools for structured history-taking and diagnostic support, especially in settings with limited vestibular expertise. Future studies should test whether such systems improve clinical decisions, reduce unnecessary resource use, and maintain safety across languages and health care settings.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/100442","type":"journal-article","created":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T04:57:23Z","timestamp":1786337843000},"page":"e100442-e100442","source":"Crossref","is-referenced-by-count":0,"title":["Conversational Large Language Models for Vestibular Diagnosis in Outpatient Clinics: Prospective Multicenter Diagnostic Accuracy Study"],"prefix":"10.2196","volume":"28","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-4641-8069","authenticated-orcid":false,"given":"Chongkai","family":"Lu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-0491-7540","authenticated-orcid":false,"given":"Ruiqi","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8646-8488","authenticated-orcid":false,"given":"Huaili","family":"Jiang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4091-1610","authenticated-orcid":false,"given":"Yanping","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3609-6804","authenticated-orcid":false,"given":"Sulin","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-3773-2480","authenticated-orcid":false,"given":"Qin","family":"Lin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4946-4140","authenticated-orcid":false,"given":"Peixia","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5338-6910","authenticated-orcid":false,"given":"Huawei","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,8,7]]},"reference":[{"issue":"3","key":"R1","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2022-064057","article-title":"Vestibular syndromes, diagnosis and diagnostic errors in patients with dizziness presenting to the emergency department: a cross-sectional study","volume":"13","author":"Comolli","journal-title":"BMJ Open"},{"issue":"12","key":"R2","doi-asserted-by":"publisher","first-page":"1465","DOI":"10.1016\/j.amjmed.2017.05.048","article-title":"Dizziness symptom type prevalence and overlap: a US nationally representative survey","volume":"130","author":"Kerber","journal-title":"Am J Med"},{"key":"R3","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-444-63437-5.00005-4","article-title":"The epidemiology of dizziness and vertigo","volume":"137","author":"Neuhauser","journal-title":"Handb Clin Neurol"},{"issue":"8","key":"R4","doi-asserted-by":"publisher","first-page":"1492","DOI":"10.1007\/s00415-014-7367-y","article-title":"Health services utilization of patients with vertigo in primary care: a retrospective cohort study","volume":"261","author":"Grill","journal-title":"J Neurol"},{"issue":"1","key":"R5","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1515\/dx-2020-0124","article-title":"Stroke hospitalization after misdiagnosis of \u201cbenign dizziness\u201d is lower in specialty care than general practice: a population-based cohort analysis of missed stroke using SPADE methods","volume":"9","author":"Chang","journal-title":"Diagnosis (Berl)"},{"issue":"9","key":"R6","doi-asserted-by":"publisher","DOI":"10.1007\/s00415-025-13291-x","article-title":"Diagnostic challenges in vestibular migraine-clinical differentiation from Meni\u00e8re\u2019s disease and discrepancies with current classification criteria","volume":"272","author":"Huppert","journal-title":"J Neurol"},{"issue":"4","key":"R7","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2024-085715","article-title":"Prospective diagnostic accuracy study of history taking and physical examination for adults with vertigo in general practice: study protocol","volume":"14","author":"Ross","journal-title":"BMJ Open"},{"issue":"10","key":"R8","doi-asserted-by":"publisher","first-page":"3092","DOI":"10.1111\/ene.15448","article-title":"Validation of a comprehensive diagnostic algorithm for patients with acute vertigo and dizziness","volume":"29","author":"Filippopulos","journal-title":"Eur J Neurol"},{"issue":"3","key":"R9","first-page":"154","volume":"95","author":"Muncie","journal-title":"Am Fam Physician"},{"issue":"3","key":"R10","doi-asserted-by":"publisher","first-page":"577","DOI":"10.1016\/j.ncl.2015.04.011","article-title":"TiTrATE: a novel, evidence-based approach to diagnosing acute dizziness and vertigo","volume":"33","author":"Newman-Toker","journal-title":"Neurol Clin"},{"issue":"1","key":"R11","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-025-01880-z","article-title":"Clinical decision support for vestibular diagnosis: large-scale machine learning with lived experience coaching","volume":"8","author":"Callejas Pastor","journal-title":"NPJ Digit Med"},{"issue":"8","key":"R12","doi-asserted-by":"publisher","DOI":"10.2196\/34126","article-title":"A questionnaire-based ensemble learning model to predict the diagnosis of vertigo: model development and validation study","volume":"24","author":"Yu","journal-title":"J Med Internet Res"},{"issue":"6","key":"R13","doi-asserted-by":"publisher","DOI":"10.3390\/healthcare13060603","article-title":"A review of large language models in medical education, clinical decision support, and healthcare administration","volume":"13","author":"Vrdoljak","journal-title":"Healthcare (Basel)"},{"issue":"1","key":"R14","doi-asserted-by":"publisher","DOI":"10.1038\/s44387-025-00011-z","article-title":"Large language models for disease diagnosis: a scoping review","volume":"1","author":"Zhou","journal-title":"NPJ Artif Intell"},{"key":"R15","doi-asserted-by":"publisher","DOI":"10.3389\/fdgth.2025.1659134","article-title":"Large language models in real-world clinical workflows: a systematic review of applications and implementation","volume":"7","author":"Artsi","journal-title":"Front Digit Health"},{"key":"R16","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.h5527","article-title":"STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies","volume":"351","author":"Bossuyt","journal-title":"BMJ"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1001\/jama.2024.21972","article-title":"World Medical Association Declaration of Helsinki: ethical principles for medical research involving human participants","volume":"333","author":"World Medical Association","journal-title":"JAMA"},{"key":"R18","unstructured":"Ethics and governance of artificial intelligence for health: guidance on large multi-modal models. World Health Organization; 2024. URL: https:\/\/iris.who.int\/server\/api\/core\/bitstreams\/e9e62c65-6045-481e-bd04-20e206bc5039\/content [Accessed 11-07-2026]"},{"key":"R19","doi-asserted-by":"publisher","DOI":"10.3389\/fneur.2023.1254080","article-title":"What is the current status of primary care in the diagnosis and treatment of patients with vertigo and dizziness in Switzerland? A national survey","volume":"14","author":"Zwergal","journal-title":"Front Neurol"},{"issue":"5","key":"R20","doi-asserted-by":"publisher","first-page":"442","DOI":"10.1111\/acem.14728","article-title":"Guidelines for reasonable and appropriate care in the emergency department 3 (GRACE-3): acute dizziness and vertigo in the emergency department","volume":"30","author":"Edlow","journal-title":"Acad Emerg Med"},{"issue":"3","key":"R21","doi-asserted-by":"publisher","first-page":"788","DOI":"10.1161\/STROKEAHA.117.016979","article-title":"Diagnosing stroke in acute dizziness and vertigo: pitfalls and pearls","volume":"49","author":"Saber Tehrani","journal-title":"Stroke"},{"issue":"3","key":"R22","doi-asserted-by":"publisher","first-page":"403","DOI":"10.1177\/0194599816689660","article-title":"Clinical practice guideline: benign paroxysmal positional vertigo (update) executive summary","volume":"156","author":"Bhattacharyya","journal-title":"Otolaryngol Head Neck Surg"},{"issue":"3","key":"R23","doi-asserted-by":"publisher","first-page":"244","DOI":"10.1001\/jamaneurol.2022.4944","article-title":"Effect of self-treatment of recurrent benign paroxysmal positional vertigo: a randomized clinical trial","volume":"80","author":"Kim","journal-title":"JAMA Neurol"},{"issue":"10","key":"R24","doi-asserted-by":"publisher","first-page":"e537","DOI":"10.1016\/S2589-7500(20)30218-1","article-title":"Reporting guidelines for clinical trial reports for interventions involving artificial intelligence: the CONSORT-AI extension","volume":"2","author":"Liu","journal-title":"Lancet Digit Health"},{"key":"R25","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-020-0221-y","article-title":"An overview of clinical decision support systems: benefits, risks, and strategies for success","volume":"3","author":"Sutton","journal-title":"NPJ Digit Med"},{"issue":"2","key":"R26","doi-asserted-by":"publisher","DOI":"10.1162\/99608f92.5317da47","article-title":"How Is ChatGPT\u2019s behavior changing over time?","volume":"6","author":"Chen","journal-title":"Harv Data Sci Rev"},{"issue":"1","key":"R27","doi-asserted-by":"publisher","DOI":"10.1186\/s13012-025-01445-4","article-title":"The implementation challenge of computerised clinical decision support systems for the detection of disease in primary care: systematic review and recommendations","volume":"20","author":"Derksen","journal-title":"Implement Sci"},{"issue":"2","key":"R28","doi-asserted-by":"publisher","DOI":"10.1136\/bmjgh-2019-002067","article-title":"Electronic clinical decision support algorithms incorporating point-of-care diagnostic tests in low-resource settings: a target product profile","volume":"5","author":"Pell\u00e9","journal-title":"BMJ Glob Health"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","deposited":{"date-parts":[[2026,8,10]],"date-time":"2026-08-10T04:57:25Z","timestamp":1786337845000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2026\/1\/e100442"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,7]]},"references-count":28,"URL":"https:\/\/doi.org\/10.2196\/100442","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,7]]},"article-number":"v28i3e100442"}}